In-vitro comparison of LC-DCP- and LCP-constructs in the femur of newborn calves – a pilot study
- 3.6k Downloads
To compare the biomechanical in-vitro characteristics of limited-contact dynamic compression plate (LC-DCP) and locking compression plate (LCP) constructs in an osteotomy gap model of femoral fracture in neonatal calves. Pairs of intact femurs from 10 calves that had died for reasons unrelated to the study were tested. A 7-hole LC-DCP with six 4.5 mm cortical screws was used in one femur and a 7-hole LCP with four 5.0 mm locking and two 4.5 mm cortical screws was used in the corresponding femur. The constructs were tested to failure by cyclic compression at a speed of 2 mm/s within six increasing force levels.
The bone-thread interface was stripped in 21 of 80 cortical screws (26.3%) before a pre-set insertion torque of 3 Nm was achieved. Only 3 corresponding intact pairs of constructs could be statistically compared for relative structural stiffness, actuator excursion and width of the osteotomy gap. Relative structural stiffness was significantly greater, actuator excursion and width of the osteotomy gap were significantly smaller in the LCP constructs. While failure occurred by loosening of the screws in the LC-DCP constructs, locking constructs failed by cutting large holes in the soft distal metaphyseal bone.
An insertion torque sufficient to provide adequate stability in femurs of newborn calves could not be achieved reliably with 4.5 mm cortical screws. Another limiting factor for both constructs was the weak cancellous bone of the distal fracture fragment. LCP constructs were significantly more resistant to compression than LC-DCP constructs.
KeywordsFemoral Fracture Distal Fragment Bovine Viral Diarrhoea Lock Compression Plate Cortical Screw
Fractures of the os femoris are common in newborn calves [1, 2, 3, 4, 5]. Femoral and tibial fractures rank second to metacarpal and metatarsal fractures in order of frequency of long bone fractures in cattle [6, 7]. The most common cause of femoral fractures in calves is excessive traction during delivery, but trauma, such as the dam standing on the calf, and bovine viral diarrhoea (BVD) virus infection are other causes [8, 9].
Conservative treatment of femoral fractures is rarely successful [2, 3, 4, 15]. When a Thomas splint  or intramedullary pins  were used for fixation, the outcome was significantly worse for fractures in the distal metaphysis compared with fractures of the diaphysis. Fixation of distal metaphyseal femoral fractures using Steinmann pins was associated with pin migration and instability [3, 5, 11]. In contrast, the outcome after plate osteosynthesis was not affected by the location of the fracture [2, 16]. Recently, a novel intramedullary interlocking nail was used for repair of femoral fractures in newborn calves with a good prognosis regardless of the location of the fracture .
The softness of femoral bone is one of the major concerns of plate osteosynthesis in neonatal calves [1, 2, 13]. Soft neonatal bone can predispose to loosening of the screws and subsequent instability of the fixation. In human medicine, angle-stable implants are used for osteosynthesis in soft and osteoporotic bones. The aim of the present in-vitro study was to compare a limited-contact dynamic compression plate (LC-DCP) construct and a locking compression plate (LCP) construct in an osteotomy-gap model of femoral fractures in newborn calves.
Before creation of the bone defect, a broad 7-hole LC-DCP was used in one femur and a 7-hole LCP was used in the contralateral femur of the same animal. The plates were contoured to the craniolateral surface of the bones. They were placed as far distally as possible adjacent to the femoropatellar joint pouch. Six 4.5 cortical screws were inserted with the LC-DCP plate. For the LCP, two 4.5 mm cortical screws were placed closest to the defect, and 5.0 mm locking screws were used in the four remaining peripheral holes. The screws were not tightened to the pre-set final torque. In each bone of every pair, a 12-mm osteotomy defect was created with an oscillating saw at the distal aspect of the femoral diaphysis. The screws were finally tightened using a screw driver with adjustable torque (torque screw driver, Hoffmann Group, Munich, Germany). Lines were drawn with a waterproof marker from the screw heads to the plate surface and additional lines were drawn on the bones along the border of the plates. These were used to identify rotation and migration of the screws in the plate holes and displacement of the plate. The bone-plate constructs were mounted in a servo-hydraulic testing machine (Bionix 858, MTS Systems, Minneapolis, USA; Figure 2) and tested at axial compression using 100 loading-relaxation cycles per testing level. The maximum force was increased after each testing level, from 500 N in the first stage to 1000 N in the second stage followed by four subsequent steps of 250 N each to a maximum force of 2000 N. The axial excursion of the actuator of the testing machine was recorded at cycle 1, 50 and 100 during each level. The number of test cycles to failure was recorded for each construct. Failure was defined as bone-to-bone contact between fragments. After failure, the constructs were examined and the plates removed to record the extent of drill-hole deformation and the number of permanently deformed screws.
For descriptive data analysis of all constructs, SPSS® 16.0 (SPSS Inc., Illinois, Chicago, USA) and Microsoft® Office Excel 2003 (Microsoft Corporation, Redmond, Washington, USA) were used. Frequencies were compared with Chi-squared tests or Fisher’s exact tests for differences between the two constructs. SAS® 9.1 (SAS Institute Inc., Cary, North Carolina, USA) was used to correct the data for repeated measures within a bone pair, and the animal from which the bones originated was defined as a random effect (PROC Mixed procedure). P ≤ 0.05 was considered statistically significant.
Stripping of the bone threads of the screw hole before the predetermined torque of 3 Nm had been reached occurred during insertion of 21 of 80 (26.3%) cortical screws; 19 of these were located in the LC-DCP and 2 in the LCP constructs (p < 0.001). Tightening of all screws of a construct with the defined torque was only achieved in three LC-DCP and eight LCP constructs (p = 0.07; Fisher’s exact test). Only the three intact corresponding constructs in which the screws could be tightened to the pre-set torque were statistically analysed with respect to relative structural stiffness, maximum axial excursion and width of the osteotomy gap. However, all 20 constructs were tested until failure and then compared with descriptive statistics.
Maximum excursion (in mm, median values for each construct) of the actuator for test cycle 5, 50 and 100
Loading cycle 5
Loading cycle 50
Loading cycle 100
Maximal changes of gap width (calculated from the median values) after the first 4 test levels
Maximum width change (in percent)
The axial compression testing led to craniolateral rotational movement (Figure 2) of the distal fragment in all the constructs. This deviation in the axis appeared early in the testing cycles and was more pronounced in the LC-DCP constructs. Screw No. 6 appeared to be the centre of this rotational movement based on the observation that there were only minor changes in the appearance of screw hole No. 6, and more severe changes at screw holes No. 5 and 7. The drill holes did not have major macroscopic changes in the trans-cortex and proximal fragment after removal of all constructs. In the cis-cortex of the distal fragment, the drill holes were markedly widened, mostly in a horizontal direction.
During the experiment, 32 of 120 (26.6%) screws underwent permanent plastic deformation under loading; all screws were from the distal fracture fragment. In total, 25 of the 80 cortical (31.3%) and seven of the 40 locking screws (17.5%) were affected (p = 0.108). The latter were deformed at the transition from screw head to body, while the cortical screws were deformed at the transition from screw head to body or in the middle of the body. Cortical screw No. 5 underwent permanent plastic deformation in seven of 10 constructs in both groups.
To the authors’ knowledge, there are no published guidelines regarding the optimum insertion torque for cortical screws in the calf femur. Although a preliminary trial established an insertion torque of 3 Nm for 4.5 mm cortical screws, a large number of cortical screws caused stripping of the thread in the soft bone of the distal metaphysis and epiphysis before the pre-set value was reached. This discrepancy is believed to mainly reflect the relative weakness of calf bones  and to a much lesser extent the differences in the structures of the femoral bones among individual calves. An insertion torque greater than 3 Nm is necessary to achieve adequate stability and to prevent movement between the components of an osteosynthesis [18, 19]. The highest torque achievable in osteoporotic bone is generally considered to be 3 Nm [18, 20, 21, 22]. It can therefore be concluded that the femoral metaphyses of newborn calves are as weak as osteoporotic bone. As a consequence of this weakness, no cortical screw in the distal fragment of the LC-DCP group remained firmly fixed after the second testing level. The number of loosened screws was much higher in the LC-DCP constructs which was associated with earlier implant failure. In clinical cases, stripped cortical screws are commonly replaced with 6.5 mm cancellous screws [2, 18]. However, cortical screws were chosen in the present study because cancellous screws did not increase the fixation strength of an LC-DCP construct in an osteoporosis model [23, 24]. In addition, in a study that investigated the holding power of different screws, there were no significant differences among the holding power of 4.5 mm and 5.5 mm cortical screws and 6.5 mm cancellous screws in the diaphysis and metaphysis of calf femurs . The insertion torque of cancellous screws in the metaphysis and epiphysis of calf femurs should be investigated in further studies.
The transition from the diaphysis to the distal metaphysis of the femur was chosen for the experimental fracture site because this is a common location of fractures in newborn calves. Furthermore, plate fixation at this location was thought to be more suitable than fixation using an intramedullary pin, whereas in the diaphysis either technique may be used [3, 11]. A relatively large osteotomy gap of 12 mm was chosen to ensure that the plate, rather than the fracture fragments, was the load-transferring component of the osteosynthesis [21, 26, 27, 28]. Such a large osteotomy gap is not usually seen in clinical cases and was probably a main factor in the rapid screw loosening in the LC-DCP group of the present study.
Permanent plastic deformation occurred in almost twice as many cortical screws than locking screws, which underlines the greater strength of the locking screws. In the locking screws, bending occurred more commonly at the transition of the body to the head of the screw. In contrast, bending in cortical screws occurred at the head or at the body. Interestingly, the proportions of bent screws in the No. 5 position were the same for LCP and LC-DCP constructs after failure. This means that the peripherally-placed locking screws could protect the cortical screw in the more central position only during the initial testing stages. Conventional cancellous screws at the plate ends may be considered in an attempt to reduce stress concentration  in calf femurs.
The initial load of 500 N corresponds to a weight of approximately 50 kg, which corresponds to the weight of a newborn calf. However, it can be assumed that the limbs of a newborn calf that struggles to stand for the first time undergo loads that correspond to multiples of their bodyweight. In a study involving human femurs, a load corresponding to three times the bodyweight was used to test different fixation systems .
Relative structural stiffness was significantly greater in the LCP constructs than in the LC-DCP constructs. Likewise, maximum axial excursion and width of the osteotomy gap were significantly smaller in LCP in a validated model of the osteoporotic femoral diaphysis .
The distal position of the plate caused bridging of the growth plate by the two most distal screws. This was inevitable in the LCP because of the predetermined entry angle of the locking screws, whereas it could have been partially prevented in the LC-DCP constructs by changing the entry angle of the cortical screws. However, for better comparison of the two constructs, similar insertion angles were used in all screws. Bridging of the growth plates by screws should be avoided in clinical cases, but in femoral fractures close to the distal metaphysis this is not always possible without severely compromising stability in large animals. In such cases, the implants should be removed as soon as possible once healing of the fracture is complete [2, 19, 32].
In large animal long-bone fractures, the use of two plates in a 90-degree configuration is recommended . In an experimental study that compared different implants, osteosynthesis with two LCPs was better than osteosynthesis with two DCPs, LC-DCPs or clamp rod internal fixation systems for the fixation of a simple oblique fracture in an equine long-bone model . Because the interaction of two plates would have complicated the interpretation of the findings, we chose to use only one plate in our model. However, in a clinical situation, a second plate could have prevented the rotational deviation.
An insertion torque sufficient to provide adequate stability for plate fixation in femurs of newborn calves could not be achieved reliably with 4.5 mm cortical screws. Femoral metaphyses of the calves studied were as weak as human osteoporotic bone. Cortical screws in the metaphyses should be replaced by cancellous screws in further studies. A main limiting factor for the stability of both LC-DCP and LCP constructs was the distal fracture fragment. However, relative structural stiffness was significantly greater and actuator excursion and width of the osteotomy gap significantly smaller in the LCP constructs.
The authors thank the University Bayern e. V. for the postgraduate scholarship for M. Hoerdemann and the Clinic for Ruminants, Veterinary Faculty, Ludwig-Maximilians-University Munich, Germany for financial support.
- 2.Nuss K, Spiess A, Feist M, Köstlin R: Treatment of long bone fractures in 125 newborn calves - a retrospective study. Tieraerztl Prax. 2011, 39 (G): 15-26.Google Scholar
- 6.Tulleners E: Management of bovine orthopedic problems. Part I. Fractures. Comp Cont Ed. 1986, 8 (2): 69-80.Google Scholar
- 7.Koestlin RG, Nuss K, Elma E: Metacarpal and metatarsal fractures in cattle. Treatment and results. Tieraerztl Prax. 1990, 18 (2): 131-144.Google Scholar
- 16.Nemeth F: Treatment of supracondylar fractures of the femur in large animals. In 12 th World Buiatrics Congress. Amsterdam: The Netherlands; 1982:791–793.Google Scholar
- 19.Bramlage LR: Management of longbone fractures in the foal. Large An Pract. 1983, 5 (2): 77-82.Google Scholar
- 21.Korner J, Diederichs G, Arzdorf M, Lill H, Josten C, Schneider E, Linke B: A biomechanical evaluation of methods of distal humerus fracture fixation using locking compression plates versus conventional reconstruction plates. J Orthop Trauma. 2004, 18 (5): 286-293. 10.1097/00005131-200405000-00004.CrossRefPubMedGoogle Scholar
- 25.Kirpensteijn J, Roush J, St Jean G, DeBowes R, Gaughan E: Holding Power of Orthopaedic Screws in Femora of Young Calves. Vet Comp Orthop Traumatol. 1993, 6: 16-20.Google Scholar
- 28.Goh CS, Santoni BG, Puttlitz CM, Palmer RH: Comparison of the mechanical behaviors of semicontoured, locking plate-rod fixation and anatomically contoured, conventional plate-rod fixation applied to experimentally induced gap fractures in canine femora. Am J Vet Res. 2009, 70 (1): 23-29. 10.2460/ajvr.70.1.23.CrossRefPubMedGoogle Scholar
- 30.Heiney JP, Barnett MD, Vrabec GA, Schoenfeld AJ, Baji A, Njus GO: Distal femoral fixation: a biomechanical comparison of trigen retrograde intramedullary (i.m.) nail, dynamic condylar screw (DCS), and locking compression plate (LCP) condylar plate. J Trauma. 2009, 66 (2)): 443-449.CrossRefPubMedGoogle Scholar
- 34.Florin M, Arzdorf M, Linke B, Auer JA: Assessment of stiffness and strength of 4 different implants available for equine fracture treatment: a study on a 20 degrees oblique long-bone fracture model using a bone substitute. Vet Surg. 2005, 34 (3): 231-238. 10.1111/j.1532.950X.2005.00035.x.CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.