- Email: [email protected]
Biomechanics of femoral interlocking nails
Injury, Int. J. Care Injured 31 (2000) 437±443
www.elsevier.com/locate/injury
Biomechanics of femoral interlocking nails P. Schandelmaier*, O. Farouk, C. Krettek, N. Reimers, J. Mannû, H. Tscherne Trauma Department, Hannover Medical School, Carl Neuberg Str 1, 30623, Hannover, Germany Accepted 9 February 2000
Abstract Today there is a variety of dierent interlocking intramedullary nail designs available for the femur. We compared dierent nail types in the bone implant complex (BIC) of four unreamed solid nails and a slotted reamed nail with simulated comminuted mid shaft fractures to see if there are major dierences in stiness for axial load, bending and torsion. The fractures were simulated by a 2 cm defect osteotomy in paired human cadaver femora. Each bone was tested intact in a universal testing machine, osteotomy and osteosynthesis were performed, and the BIC was tested. Relative stiness was calculated. In torque testing the unslotted solid nail showed signi®cantly more stiness (0.6±1.8 Nm/8) compared to the slotted nail (0.2 Nm/8). Compared to intact bone (6.9 Nm/8), both groups of nails were signi®cantly less sti (relative stiness 2±20%). In axial load and bending testing the large diameter unreamed nail showed signi®cantly higher stiness (32±68%). This study shows that stiness of bone implant complex in interlocking femoral nails is more dependent on nail pro®le than on the press®t of nails in the medullary canal. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: Biomechancis; Fracture-®xation-intramedullary; Femoral-fractures-diaphyseal; Unreamed; Cadaver-study-human
1. Introduction The ®rst interlocking nail as proposed by KuÈntscher was based on a slotted nail with low torque stiness [10]. In a locking nail, the connection between the nail and the bone is not created by elastic jamming in the medullary cavity, but by a screw connection or a bolt. Even subsequent locking nails developed by Klemm and Schellmann (K&S) [8], Grosse and Kempf (G&K) [7], BoÈrner and Mattheck (B-M) [11], and AO Universal (AOU) were slotted. However, unslotted nails were also developed, namely Russell and Taylor Delta (RTD) [1], Russell and Taylor Reconstruction (RTR) [1], Brooker and Wills (B&W) [2], as well as the AO unreamed femoral nail (UFN) [9]. Studies of the stiness of various implants within the bone implant complex (BIC) have been performed on the femur, the * Corresponding author. Tel.: +49511/532-2099; fax: +49511/5325877. E-mail address: [email protected] (P. Schandelmaier).
tibia and the humerus [4,6,13], but there have been none concerning solid and slotted nail pro®les in the femur so far. The aim of this study was to compare dierent solid femoral nails to a slotted nail with special attention on the in¯uence of nail diameter on the stiness of the BIC. 2. Materials and methods Twenty pairs of cold-preserved human femora of 400 2 10 mm in length were used. The femora were taken during autopsy from donors with healthy bones (donor age 22±55 years). The surrounding soft tissue was removed, the bones were wrapped in moist cloths and stored at ÿ188C until thawed 24 h prior to testing. Anterior±posterior and lateral radiographs were taken to rule out pathological changes, then the isolated bones were embedded in a cast frame (Vel-Mix-Stone, Kerr Co., Karlsruhe, Germany) (Fig. 1). The fracture model was tested in a mechanically un-
0020-1383/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 0 ) 0 0 0 1 9 - X
438
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
favourable situation, a comminuted fracture. The fracture model was created by an osteotomy 10 mm proximally and 10 mm distally of the middle of the femoral shaft and removal of the resultant fragment. Prior to osteotomy, the medullary cavity was opened with an awl in the axis of the femoral shaft near the piriformis fossa. All implants were implanted in accordance with the manufacturers' recommendations. In the study, all types of nails were proximally and distally ®xed with screws, the diameter varied between 9, 11 and 12 mm. The following implants were used: AO Universal Femoral Nail (AOU) 11 mm, AO Unreamed Femoral Nail (steel) 9 mm, locked with 3.2±3.9 mm bolts (SUFN3.9), SUFN 9 mm with 4.3± 4.9 mm bolts (SUFN4.9), AO unreamed Femoral Nail (titanium) 12 mm (12TUFN), locked with 4.3±4.9 mm bolts, TUFN 9 mm, locked with 4.3±4.9 mm bolts (9TUFN) (Table 1). After preparation of the medullary cavity, the nails were driven into place. In the case of the reamed AOU, the AO reaming instrumentation (Synthes Mathys., Bochum, Germany) was used; the nail was
then driven in over the guide wire. Proximal and distal locking was carried out using the free-hand technique with an image intensi®er. The BIC was embedded into the cast frame so that the femoral axis stood vertically in the middle of the base of the cement block. All tests were performed with a ZWICK 1445 universal testing machine (UTM) (Zwick Co., Ulm, Germany). In order to create torsional and axial loading, the BIC was positioned proximally in a custom-made aluminium clamp. Distally, the BIC was clamped in over the cast anchorage block. Thereby, femoral axis and load axis were aligned. For axial testing, a pre-load of 5 N and a maximal load of 1100 N was applied, then reduced to 0 N whilst recording the distance; the test velocity was 1.5 mm/min (Fig. 1). The torsion tests were conducted as a superimposed torsion/pressure test with a constant axial load of 10 N and a maximum moment of 4 Nm in both directions; the pre-moment was +0.05 Nm and the test velocity 188/min. The maximum loads chosen were such that the tests did not cause fracturing of implants or bones. No screws broke during testing. The bending load was carried out using a four-point bending test system (Fig. 2). The bending distance was 127 mm and the support distance 380 mm; the force was applied via steel rollers of 12.7 mm in diameter. With a pre-moment of 0.05 Nm, a load of up to 66 Nm was applied at a test velocity of 40 mm/min. Prior to testing the bones in a BIC, each bone was tested intact, so relative values for stiness could be calculated. The bending moment and bending angle were calculated from the transverse distance and the eective force acting on the known dimensions of the transverse beam. Torsional-stiness was de®ned as Load/Degree angle-deformation (Nm/8) between both ending points at +4 Nm and ÿ4 Nm. The resilience rate of the BIC between 100 and 1000 N in N/mm served as the parameter for axial stiness of the BIC. For purposes of evaluation, the bending stiness was de®ned in simpli®ed terms as the moment per degree of bending angle (Nm/8) between 100 and 2000 Nm. Between 0 and 100 Nm the implants showed `play' in the medullary cavity. Therefore, this range was not included in the evaluation. The data recorded on the stiness of the BIC was compared with the torsional stiness of each corresponding intact bone. 3. Results
Fig. 1. Torque testing being carried out with the universal testing machine (ZWICK 1445). The BIC is positioned proximally in a custom made aluminum clamp. Distally, the BIC was clamped in over the cast anchorage block.
Compared to the stiness of the intact bone, stiness of the BIC was between 2 and 20% (Table 2). Between the solid unslotted nails, no signi®cant dierences were observed. The isolated bone without osteotomy showed the highest amount of stiness, being ®ve
Short form AOU
SUFN3.9
SUFN4.9
9TUFN
12TUFN
Implant
AO universal femoral nail
AO Unreamed femoral nail 9 mm Steel
AO Unreamed femoral nail 9 mm Steel
AO Unreamed femoral nail 9 mm Titanium
AO Unreamed femoral nail 12 mm Titanium
Table 1 Tested implants
12
9
9
9
11
Diameter (mm) 1.2
±
±
±
±
Pro®le thickness (mm)
Solid, round, ¯uted
Solid, round
Solid, round
Solid, round
Tube slotted clover leaf
Type of pro®le
Pro®le
0
0
0
0
2.6
Width of slot (mm)
unreamed
unreamed
unreamed
unreamed
11.5
Reamed up to (mm)
Material
Synthes Mathys GmbH, D44791 Bochum
Synthes Mathys GmbH, D44791 Bochum
Synthes Mathys GmbH, D44791 Bochum
Distributor
4.3/4.9 mm Titanium Alloy (Ti± Synthes Mathys GmbH, DAL6±Nb7) 44791 Bochum
4.3/4.9 mm Titanium Alloy (Ti± Synthes Mathys GmbH, DAL6±Nb7) 44791 Bochum
4.3/4.9 mm Stainless Steel
3.2/3.9 mm Stainless Steel
4.3/4.9 mm Stainless Steel
Distal locking
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 439
440
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
times as high as the stiness of the 12TUFN-BIC and 50 times as high as the slotted AOU-BIC. In contrast to the torsion testing, there were only slight dierences between the implants, with the 12TUFN and SUFN4.9 showing a signi®cant higher stiness compared to the rest of the tested implants. The respective stiness was between 41 and 65% of the intact femur (Table 3). Overall, the bending stiness spanned a relatively small range of 7±12 Nm/8 in comparison with the torsional range of 0.18±1.76 Nm/8. The isolated, nonosteotomized bone, however, was signi®cantly more rigid, registering 24 Nm/8 torque stiness (Table 4). 4. Discussion Knowledge about the biomechanical characteristics of femur locking nails is still rather limited. The in¯uence of load on a human femur and a femoral nail were measured telemetrically by means of a instrumented femoral nail [14]. Even after 6 months torsionloads up to 3.1 Nm could be measured on an implanted 16-mm AO Universal nail, the highest load being recorded standing on one limb. The torsion-stiness of the nail-pro®le is of less interest in unlocked nails, because without locking in a fracture lacking cortical support only a slight amount of torsion-stiness can be introduced [15]. Grundnes and Reikeras [3] compared three dierent kinds of medullary nails in the femora of rats. In the ®rst group, osteotomies were treated with a rotationally and axially rigid nail, comparable to a large diameter unslotted locking nail with
thick shaft screws. In the second group the nail was rotationally rigid and axially non-rigid, comparable to an unslotted nail with a thin shaft screw or dynamic locking (comparable in our series to a UFN). In the third group the nail was unstable in both planes, clinically best comparable to a dynamized, slotted nail. The authors reported a higher bending stiness, a higher bending moment up to fracture, and a greater amount of bone-absorbed energy up to fracture in those animals implanted with a medullary nail of high torsional stiness. Molster [12] observed that in rats whose femoral osteotomies were treated with medullary nails of varying stiness, the group with higher torsional stiness of the implant achieved a higher stiness of the bone signi®cantly earlier than the group with lower torsional stiness. In femoral osteotomies in dogs, the author observed a higher breaking load and stiness in dogs implanted with unslotted medullary nails compared with those implanted with slotted nails [16]. In our series the model of a comminuted fracture was used, with both of the main fragments lacking any lateral support. Therefore statements can only be made about the primary stability in a BIC achieved by means of interlocking, representing a worst case scenario. Conclusions can be drawn about the reaction under load of the femur stabilized with an osteosynthesis in the early postoperative period. Testing for fatigue and alteration was not undertaken in this study. The original stiness and breaking load of the intact femur can not be matched by any of the BIC tested, neither for rotation, axial load nor bending. The stiness achieved by the BIC is in¯uenced by the diameter
Fig. 2. Four point bending with the universal testing machine.
p < 0.01 ANOVA LSD test.
15712199 13642257 17522143 6232145 665245 3729293
AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
a
BIC stiness (N/mm)
Axial stiness Implant
Table 3 Axial stiness 100±1000 N
p < 0.01 ANOVA LSD test.
0.1820.04 1.4420.15 1.7620.19 0.5820.08 0.8520.08 6.8620.28
AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
a
BIC stiness (Nm/8)
Torque stiness Implant
4926 4127 6225.5 50221 6525
BIC relative stiness % intact bone2SEM
220.4 1621.6 2021.8 1221.5 1821.8
BIC relative stiness % intact bone2SEM
Table 2 Torque stiness of BIC and isolated nail 5 Nm to ÿ5 Nm with standard error of each BIC
2684229 3573261 51382155 23162156 30832145
Nail stiness (N/mm)
0.0720.0004 1.4420.03 2.3720.06 1.7820.13 2.6820.12
Nail stiness (Nm/8) ±
±
SUFN3.9
±
SUFN3.9
a
±
AOU
a
AOU
±
SUFN4.9
±
SUFN4.9
±
9TUFN
±
9TUFN
±
±
12TUFN
12TUFN
±
Intact femur
±
Intact femur
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 441
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
±
12TUFN
32.223.1 47.222.9 46.223.1 4922.9 58.324.1
±
a
±
±
±
9TUFN SUFN4.9
References
p < 0.01 ANOVA LSD test. a
7.1620.6 7.5520.3 7.0520.2 7.3520.6 11.7921.3 24.125.8 AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
3723.5 4421.5 4321.5 4523 6827
SUFN3.9 AOU Nail stiness (Nm/8) BIC stiness (Nm8)
BIC relative stiness % intact bone2SEM
5. Conclusions 1. It is the pro®le which is decisive for the torsional stiness of femoral locking nails in the BIC. The presence of a slot in the pro®le is of special importance. 2. Unslotted nails have a signi®cantly higher torsional stiness than slotted nails.
Bending Stiness Implant
Table 4 Bending Stiness
and pro®le of the implants. Slotted nails show only 2% of the torsional stiness achieved by unslotted medullary nails, although the pro®le of both nails is almost identical. Unslotted solid implants showed no signi®cant dierences. The diameter is of the uttermost importance for bending-stiness, in¯uencing the bending-stiness in the 4th potency. In case of calculating the pro®le-stiness, implants with a strength of 11 and 12 mm should demonstrate the highest stiness [5]. The implants differ only slightly when compared in the BIC (stiness being 32±58% of the intact bone) to the testing for torsional-stiness. This fact might be explained by the small dierences in diameter of the implants. While testing for axial stiness, only the SUFN4.9 (6%) and the 12TUFN (65%) showed a signi®cant higher amount of stiness, the other implants diered only slightly (41±50%). The design of femur locking nails was derived from unlocked nails, with emphasis on an uncomplicated implantation. Although it could be assumed that locking-nails pose dierent requirements, in humans the in¯uence of the implant on fracture healing (like reducing cortical blood-¯ow during reaming) and the eectiveness of the mechanical qualities during implantation have to be recognized.
±
Intact femur
442
[1] Beals N, Durham G, Lynch G. Charakteristische mechanische Merkmale intramedullaÈrer VerriegelungsnaÈgel. Memphis, TN: Smith & Nephew Richards, 1990. [2] Brooker Jr AF, Brumback RJ. Brooker-Wills nails in treatment of infra-isthmal injuries of the femur. Journal of Trauma 1988;28:688±91. [3] Grundnes O, Reikeras O. Eects of instability on bone healing. Acta Orthop Scand 1993;64:55±8. [4] Henley BM, Monroe M, Tencer AF. Biomechanical comparison of methods of ®xation of a midshaft osteotomy of the humerus. Journal of Orthopaedic Trauma 1991;5(1):14±20. [5] Holzmann G, Meyer H, Schumpich G. Technische Mechanik Teil 3: Festigkeitslehre. Stuttgart: Teubner, 1972. [6] Johnson KD, Tencer AF, Sherman MC. Biomechanical factors aecting fracture stability and femoral bursting in closed intramedullary nailing of femoral shaft fractures, with illustrative case presentations. Journal of Orthopaedic Trauma 1987;1:1±11.
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 [7] Kempf I, Grosse A, Laorgue D. L'apport du verrouillage dans l'enclouage centro-meÂdullaire des os longs. Revue de Chirurgie orthopeÂdique 1978;64:635±51. [8] Klemm K, Schellmann WD, Vittali HP. [Intramedullary nail bolted to the femur and tibia]. Bulletin de la Societe Internationale de Chirurgie 1975;34:93±6. [9] Krettek C, Rudolf J, Schandelmaier P, et al. Unreamed intramedullary nailing of femoral shaft fractures:operative technique and early clinical experience with standard locking option. Injury 1996;27:233±54. [10] KuÈntscher G. Die Marknagelung des TruÈmmerbruchs. Langenbecks Archiv fuÈr Chirurgie 1968;322:1063±9. [11] Mockwitz J, BoÈrner M, Soldner E. Indikationen fuÈr die Verriegelungsnagelung am Unterschenkel bei TruÈmmer-, StuÈckund Etagenfrakturen. Hefte zur Unfallheilkunde 1983;161:110± 7.
443
[12] Molster AO, Gjerdet NR, Alho A, et al. Fracture healing after rigid intramedullary nailing in rats. Acta Orthop Scand 1983;54:366±73. [13] Schandelmaier P, Krettek C, Tscherne H. Biomechanical study of nine dierent tibia locking nails. Journal of Orthopaedic Trauma 1996;10:37±45. [14] Schneider E, Michel MC, Genge M, et al. In vivo load measurements by means of an instrumented interlocking nail. Journal of Biomechanics 1991;24:463. [15] Tencer AF, Johnson KD, Johnston DW, et al. A biomechanical comparison of various methods of stabilization of subtrochanteric fractures of the femur. Journal of Orthopaedic Research 1984;2:297±305. [16] Woodard PL, Self J, Calhoun J, et al. The eect of implant axial and torsional stiness on fracture healing. Journal of Orthopaedic Trauma 1988;1(4):331±40.
www.elsevier.com/locate/injury
Biomechanics of femoral interlocking nails P. Schandelmaier*, O. Farouk, C. Krettek, N. Reimers, J. Mannû, H. Tscherne Trauma Department, Hannover Medical School, Carl Neuberg Str 1, 30623, Hannover, Germany Accepted 9 February 2000
Abstract Today there is a variety of dierent interlocking intramedullary nail designs available for the femur. We compared dierent nail types in the bone implant complex (BIC) of four unreamed solid nails and a slotted reamed nail with simulated comminuted mid shaft fractures to see if there are major dierences in stiness for axial load, bending and torsion. The fractures were simulated by a 2 cm defect osteotomy in paired human cadaver femora. Each bone was tested intact in a universal testing machine, osteotomy and osteosynthesis were performed, and the BIC was tested. Relative stiness was calculated. In torque testing the unslotted solid nail showed signi®cantly more stiness (0.6±1.8 Nm/8) compared to the slotted nail (0.2 Nm/8). Compared to intact bone (6.9 Nm/8), both groups of nails were signi®cantly less sti (relative stiness 2±20%). In axial load and bending testing the large diameter unreamed nail showed signi®cantly higher stiness (32±68%). This study shows that stiness of bone implant complex in interlocking femoral nails is more dependent on nail pro®le than on the press®t of nails in the medullary canal. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: Biomechancis; Fracture-®xation-intramedullary; Femoral-fractures-diaphyseal; Unreamed; Cadaver-study-human
1. Introduction The ®rst interlocking nail as proposed by KuÈntscher was based on a slotted nail with low torque stiness [10]. In a locking nail, the connection between the nail and the bone is not created by elastic jamming in the medullary cavity, but by a screw connection or a bolt. Even subsequent locking nails developed by Klemm and Schellmann (K&S) [8], Grosse and Kempf (G&K) [7], BoÈrner and Mattheck (B-M) [11], and AO Universal (AOU) were slotted. However, unslotted nails were also developed, namely Russell and Taylor Delta (RTD) [1], Russell and Taylor Reconstruction (RTR) [1], Brooker and Wills (B&W) [2], as well as the AO unreamed femoral nail (UFN) [9]. Studies of the stiness of various implants within the bone implant complex (BIC) have been performed on the femur, the * Corresponding author. Tel.: +49511/532-2099; fax: +49511/5325877. E-mail address: [email protected] (P. Schandelmaier).
tibia and the humerus [4,6,13], but there have been none concerning solid and slotted nail pro®les in the femur so far. The aim of this study was to compare dierent solid femoral nails to a slotted nail with special attention on the in¯uence of nail diameter on the stiness of the BIC. 2. Materials and methods Twenty pairs of cold-preserved human femora of 400 2 10 mm in length were used. The femora were taken during autopsy from donors with healthy bones (donor age 22±55 years). The surrounding soft tissue was removed, the bones were wrapped in moist cloths and stored at ÿ188C until thawed 24 h prior to testing. Anterior±posterior and lateral radiographs were taken to rule out pathological changes, then the isolated bones were embedded in a cast frame (Vel-Mix-Stone, Kerr Co., Karlsruhe, Germany) (Fig. 1). The fracture model was tested in a mechanically un-
0020-1383/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 0 0 ) 0 0 0 1 9 - X
438
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
favourable situation, a comminuted fracture. The fracture model was created by an osteotomy 10 mm proximally and 10 mm distally of the middle of the femoral shaft and removal of the resultant fragment. Prior to osteotomy, the medullary cavity was opened with an awl in the axis of the femoral shaft near the piriformis fossa. All implants were implanted in accordance with the manufacturers' recommendations. In the study, all types of nails were proximally and distally ®xed with screws, the diameter varied between 9, 11 and 12 mm. The following implants were used: AO Universal Femoral Nail (AOU) 11 mm, AO Unreamed Femoral Nail (steel) 9 mm, locked with 3.2±3.9 mm bolts (SUFN3.9), SUFN 9 mm with 4.3± 4.9 mm bolts (SUFN4.9), AO unreamed Femoral Nail (titanium) 12 mm (12TUFN), locked with 4.3±4.9 mm bolts, TUFN 9 mm, locked with 4.3±4.9 mm bolts (9TUFN) (Table 1). After preparation of the medullary cavity, the nails were driven into place. In the case of the reamed AOU, the AO reaming instrumentation (Synthes Mathys., Bochum, Germany) was used; the nail was
then driven in over the guide wire. Proximal and distal locking was carried out using the free-hand technique with an image intensi®er. The BIC was embedded into the cast frame so that the femoral axis stood vertically in the middle of the base of the cement block. All tests were performed with a ZWICK 1445 universal testing machine (UTM) (Zwick Co., Ulm, Germany). In order to create torsional and axial loading, the BIC was positioned proximally in a custom-made aluminium clamp. Distally, the BIC was clamped in over the cast anchorage block. Thereby, femoral axis and load axis were aligned. For axial testing, a pre-load of 5 N and a maximal load of 1100 N was applied, then reduced to 0 N whilst recording the distance; the test velocity was 1.5 mm/min (Fig. 1). The torsion tests were conducted as a superimposed torsion/pressure test with a constant axial load of 10 N and a maximum moment of 4 Nm in both directions; the pre-moment was +0.05 Nm and the test velocity 188/min. The maximum loads chosen were such that the tests did not cause fracturing of implants or bones. No screws broke during testing. The bending load was carried out using a four-point bending test system (Fig. 2). The bending distance was 127 mm and the support distance 380 mm; the force was applied via steel rollers of 12.7 mm in diameter. With a pre-moment of 0.05 Nm, a load of up to 66 Nm was applied at a test velocity of 40 mm/min. Prior to testing the bones in a BIC, each bone was tested intact, so relative values for stiness could be calculated. The bending moment and bending angle were calculated from the transverse distance and the eective force acting on the known dimensions of the transverse beam. Torsional-stiness was de®ned as Load/Degree angle-deformation (Nm/8) between both ending points at +4 Nm and ÿ4 Nm. The resilience rate of the BIC between 100 and 1000 N in N/mm served as the parameter for axial stiness of the BIC. For purposes of evaluation, the bending stiness was de®ned in simpli®ed terms as the moment per degree of bending angle (Nm/8) between 100 and 2000 Nm. Between 0 and 100 Nm the implants showed `play' in the medullary cavity. Therefore, this range was not included in the evaluation. The data recorded on the stiness of the BIC was compared with the torsional stiness of each corresponding intact bone. 3. Results
Fig. 1. Torque testing being carried out with the universal testing machine (ZWICK 1445). The BIC is positioned proximally in a custom made aluminum clamp. Distally, the BIC was clamped in over the cast anchorage block.
Compared to the stiness of the intact bone, stiness of the BIC was between 2 and 20% (Table 2). Between the solid unslotted nails, no signi®cant dierences were observed. The isolated bone without osteotomy showed the highest amount of stiness, being ®ve
Short form AOU
SUFN3.9
SUFN4.9
9TUFN
12TUFN
Implant
AO universal femoral nail
AO Unreamed femoral nail 9 mm Steel
AO Unreamed femoral nail 9 mm Steel
AO Unreamed femoral nail 9 mm Titanium
AO Unreamed femoral nail 12 mm Titanium
Table 1 Tested implants
12
9
9
9
11
Diameter (mm) 1.2
±
±
±
±
Pro®le thickness (mm)
Solid, round, ¯uted
Solid, round
Solid, round
Solid, round
Tube slotted clover leaf
Type of pro®le
Pro®le
0
0
0
0
2.6
Width of slot (mm)
unreamed
unreamed
unreamed
unreamed
11.5
Reamed up to (mm)
Material
Synthes Mathys GmbH, D44791 Bochum
Synthes Mathys GmbH, D44791 Bochum
Synthes Mathys GmbH, D44791 Bochum
Distributor
4.3/4.9 mm Titanium Alloy (Ti± Synthes Mathys GmbH, DAL6±Nb7) 44791 Bochum
4.3/4.9 mm Titanium Alloy (Ti± Synthes Mathys GmbH, DAL6±Nb7) 44791 Bochum
4.3/4.9 mm Stainless Steel
3.2/3.9 mm Stainless Steel
4.3/4.9 mm Stainless Steel
Distal locking
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 439
440
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
times as high as the stiness of the 12TUFN-BIC and 50 times as high as the slotted AOU-BIC. In contrast to the torsion testing, there were only slight dierences between the implants, with the 12TUFN and SUFN4.9 showing a signi®cant higher stiness compared to the rest of the tested implants. The respective stiness was between 41 and 65% of the intact femur (Table 3). Overall, the bending stiness spanned a relatively small range of 7±12 Nm/8 in comparison with the torsional range of 0.18±1.76 Nm/8. The isolated, nonosteotomized bone, however, was signi®cantly more rigid, registering 24 Nm/8 torque stiness (Table 4). 4. Discussion Knowledge about the biomechanical characteristics of femur locking nails is still rather limited. The in¯uence of load on a human femur and a femoral nail were measured telemetrically by means of a instrumented femoral nail [14]. Even after 6 months torsionloads up to 3.1 Nm could be measured on an implanted 16-mm AO Universal nail, the highest load being recorded standing on one limb. The torsion-stiness of the nail-pro®le is of less interest in unlocked nails, because without locking in a fracture lacking cortical support only a slight amount of torsion-stiness can be introduced [15]. Grundnes and Reikeras [3] compared three dierent kinds of medullary nails in the femora of rats. In the ®rst group, osteotomies were treated with a rotationally and axially rigid nail, comparable to a large diameter unslotted locking nail with
thick shaft screws. In the second group the nail was rotationally rigid and axially non-rigid, comparable to an unslotted nail with a thin shaft screw or dynamic locking (comparable in our series to a UFN). In the third group the nail was unstable in both planes, clinically best comparable to a dynamized, slotted nail. The authors reported a higher bending stiness, a higher bending moment up to fracture, and a greater amount of bone-absorbed energy up to fracture in those animals implanted with a medullary nail of high torsional stiness. Molster [12] observed that in rats whose femoral osteotomies were treated with medullary nails of varying stiness, the group with higher torsional stiness of the implant achieved a higher stiness of the bone signi®cantly earlier than the group with lower torsional stiness. In femoral osteotomies in dogs, the author observed a higher breaking load and stiness in dogs implanted with unslotted medullary nails compared with those implanted with slotted nails [16]. In our series the model of a comminuted fracture was used, with both of the main fragments lacking any lateral support. Therefore statements can only be made about the primary stability in a BIC achieved by means of interlocking, representing a worst case scenario. Conclusions can be drawn about the reaction under load of the femur stabilized with an osteosynthesis in the early postoperative period. Testing for fatigue and alteration was not undertaken in this study. The original stiness and breaking load of the intact femur can not be matched by any of the BIC tested, neither for rotation, axial load nor bending. The stiness achieved by the BIC is in¯uenced by the diameter
Fig. 2. Four point bending with the universal testing machine.
p < 0.01 ANOVA LSD test.
15712199 13642257 17522143 6232145 665245 3729293
AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
a
BIC stiness (N/mm)
Axial stiness Implant
Table 3 Axial stiness 100±1000 N
p < 0.01 ANOVA LSD test.
0.1820.04 1.4420.15 1.7620.19 0.5820.08 0.8520.08 6.8620.28
AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
a
BIC stiness (Nm/8)
Torque stiness Implant
4926 4127 6225.5 50221 6525
BIC relative stiness % intact bone2SEM
220.4 1621.6 2021.8 1221.5 1821.8
BIC relative stiness % intact bone2SEM
Table 2 Torque stiness of BIC and isolated nail 5 Nm to ÿ5 Nm with standard error of each BIC
2684229 3573261 51382155 23162156 30832145
Nail stiness (N/mm)
0.0720.0004 1.4420.03 2.3720.06 1.7820.13 2.6820.12
Nail stiness (Nm/8) ±
±
SUFN3.9
±
SUFN3.9
a
±
AOU
a
AOU
±
SUFN4.9
±
SUFN4.9
±
9TUFN
±
9TUFN
±
±
12TUFN
12TUFN
±
Intact femur
±
Intact femur
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 441
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443
±
12TUFN
32.223.1 47.222.9 46.223.1 4922.9 58.324.1
±
a
±
±
±
9TUFN SUFN4.9
References
p < 0.01 ANOVA LSD test. a
7.1620.6 7.5520.3 7.0520.2 7.3520.6 11.7921.3 24.125.8 AOU SUFN3.9 SUFN4.9 9TUFN 12TUFN Intact femur
3723.5 4421.5 4321.5 4523 6827
SUFN3.9 AOU Nail stiness (Nm/8) BIC stiness (Nm8)
BIC relative stiness % intact bone2SEM
5. Conclusions 1. It is the pro®le which is decisive for the torsional stiness of femoral locking nails in the BIC. The presence of a slot in the pro®le is of special importance. 2. Unslotted nails have a signi®cantly higher torsional stiness than slotted nails.
Bending Stiness Implant
Table 4 Bending Stiness
and pro®le of the implants. Slotted nails show only 2% of the torsional stiness achieved by unslotted medullary nails, although the pro®le of both nails is almost identical. Unslotted solid implants showed no signi®cant dierences. The diameter is of the uttermost importance for bending-stiness, in¯uencing the bending-stiness in the 4th potency. In case of calculating the pro®le-stiness, implants with a strength of 11 and 12 mm should demonstrate the highest stiness [5]. The implants differ only slightly when compared in the BIC (stiness being 32±58% of the intact bone) to the testing for torsional-stiness. This fact might be explained by the small dierences in diameter of the implants. While testing for axial stiness, only the SUFN4.9 (6%) and the 12TUFN (65%) showed a signi®cant higher amount of stiness, the other implants diered only slightly (41±50%). The design of femur locking nails was derived from unlocked nails, with emphasis on an uncomplicated implantation. Although it could be assumed that locking-nails pose dierent requirements, in humans the in¯uence of the implant on fracture healing (like reducing cortical blood-¯ow during reaming) and the eectiveness of the mechanical qualities during implantation have to be recognized.
±
Intact femur
442
[1] Beals N, Durham G, Lynch G. Charakteristische mechanische Merkmale intramedullaÈrer VerriegelungsnaÈgel. Memphis, TN: Smith & Nephew Richards, 1990. [2] Brooker Jr AF, Brumback RJ. Brooker-Wills nails in treatment of infra-isthmal injuries of the femur. Journal of Trauma 1988;28:688±91. [3] Grundnes O, Reikeras O. Eects of instability on bone healing. Acta Orthop Scand 1993;64:55±8. [4] Henley BM, Monroe M, Tencer AF. Biomechanical comparison of methods of ®xation of a midshaft osteotomy of the humerus. Journal of Orthopaedic Trauma 1991;5(1):14±20. [5] Holzmann G, Meyer H, Schumpich G. Technische Mechanik Teil 3: Festigkeitslehre. Stuttgart: Teubner, 1972. [6] Johnson KD, Tencer AF, Sherman MC. Biomechanical factors aecting fracture stability and femoral bursting in closed intramedullary nailing of femoral shaft fractures, with illustrative case presentations. Journal of Orthopaedic Trauma 1987;1:1±11.
P. Schandelmaier et al. / Injury, Int. J. Care Injured 31 (2000) 437±443 [7] Kempf I, Grosse A, Laorgue D. L'apport du verrouillage dans l'enclouage centro-meÂdullaire des os longs. Revue de Chirurgie orthopeÂdique 1978;64:635±51. [8] Klemm K, Schellmann WD, Vittali HP. [Intramedullary nail bolted to the femur and tibia]. Bulletin de la Societe Internationale de Chirurgie 1975;34:93±6. [9] Krettek C, Rudolf J, Schandelmaier P, et al. Unreamed intramedullary nailing of femoral shaft fractures:operative technique and early clinical experience with standard locking option. Injury 1996;27:233±54. [10] KuÈntscher G. Die Marknagelung des TruÈmmerbruchs. Langenbecks Archiv fuÈr Chirurgie 1968;322:1063±9. [11] Mockwitz J, BoÈrner M, Soldner E. Indikationen fuÈr die Verriegelungsnagelung am Unterschenkel bei TruÈmmer-, StuÈckund Etagenfrakturen. Hefte zur Unfallheilkunde 1983;161:110± 7.
443
[12] Molster AO, Gjerdet NR, Alho A, et al. Fracture healing after rigid intramedullary nailing in rats. Acta Orthop Scand 1983;54:366±73. [13] Schandelmaier P, Krettek C, Tscherne H. Biomechanical study of nine dierent tibia locking nails. Journal of Orthopaedic Trauma 1996;10:37±45. [14] Schneider E, Michel MC, Genge M, et al. In vivo load measurements by means of an instrumented interlocking nail. Journal of Biomechanics 1991;24:463. [15] Tencer AF, Johnson KD, Johnston DW, et al. A biomechanical comparison of various methods of stabilization of subtrochanteric fractures of the femur. Journal of Orthopaedic Research 1984;2:297±305. [16] Woodard PL, Self J, Calhoun J, et al. The eect of implant axial and torsional stiness on fracture healing. Journal of Orthopaedic Trauma 1988;1(4):331±40.