Arch Orthop Trauma Surg DOI 10.1007/s00402-015-2252-4
TRAUMA SURGERY
Extra- vs. intramedullary treatment of pertrochanteric fractures: a biomechanical in vitro study comparing dynamic hip screw and intramedullary nail Lukas Weiser1 • Andreas A. Ruppel2 • Jakob V. Nu¨chtern1 • Kay Sellenschloh3 • Johannes Zeichen2 • Klaus Pu¨schel4 • Michael M. Morlock3 • Wolfgang Lehmann1
Received: 15 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Introduction Due to the demographic trend, pertrochanteric fractures of the femur will gain increasing importance in the future. Both extra- and intramedullary implants are used with good results in the treatment of these fractures. New, angular stable extramedullary implants promise increased postoperative stability even with unstable fractures. Additional trochanteric plates are intended to prevent secondary impaction, varisation and shortening of the fracture, as well as medialisation of the femoral shaft. The aim of this study was to perform a biomechanical comparison of both procedures regarding their postoperative stability and failure mechanisms. Materials and methods Twelve fresh-frozen human femurs were randomized into two groups based on the volumetric bone mineral density (vBMD). Standardized pertrochanteric fractures (AO31-A2.3) were generated and treated either with an angular stable dynamic hip screw (DHS) or an intramedullary nail (nail). Correct implant position and the tip–apex distance (TAD) were controlled postoperatively using X-ray. Specimens were mounted in a
servohydraulic testing machine and an axial loading was applied according to a single-leg stance model. Both groups were biomechanically compared with regard to native and postoperative stiffness, survival during cyclic testing, load to failure, and failure mechanisms. Results TAD, vBMD, and native stiffness were similar for both groups. The stiffness decreased significantly from native to postoperative state in all specimens (p \ 0.001). The postoperative stiffness of both groups varied non-significantly (p = 0.275). The failure loads for specimens treated with the nail were significantly higher than for those treated with the DHS (8480.8 ± 1238.9 N vs. 2778.2 ± 196.8 N; p = 0.008). Conclusions Extra- and intramedullary osteosynthesis showed comparable results as regards postoperative stiffness and survival during cyclic testing. Since the failure load of the nail was significantly higher in the tested AO31A2.3 fracture model, we conclude that intramedullary implants should be preferred in these, unstable, fractures.
L. Weiser and A. A. Ruppel contributed equally and therefore share first authorship.
Keywords Pertrochanteric fracture Hip fracture Sliding hip screw Dynamic hip screw Intramedullary nail Intertan nail
& Lukas Weiser [email protected]
Introduction
1
Department of Trauma-, Hand- and Reconstructive Surgery, University Medical Center Hamburg-Eppendorf, Martinistraße 52, 20246 Hamburg, Germany
2
Trauma and Orthopedic Department, Johannes Wesling Klinikum Minden, Minden, Germany
3
Institute of Biomechanics, TUHH Hamburg University of Technology, Hamburg, Germany
4
Institute of Forensic Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Pertrochanteric fractures are one of the most common injuries associated with osteoporosis. Due to the increasing population older than 65 years, these fractures will gain clinical as well as economic importance in the future [1, 2]. Both intramedullary and extramedullary implants are available for the treatment of trochanteric fractures; however, there is no gold standard. Recent studies show that the use of intramedullary nails has overtaken the use of
123
Arch Orthop Trauma Surg
extramedullary implants [3, 4]. Advantages such as better reconstituting of patients to their pre-operative state, lesser blood loss, and fewer complications have been described [5, 6]. Furthermore, cephalomedullary nails may be beneficial to the treatment of unstable and subtrochanteric fractures [4, 7]. On the contrary, it has been shown that extramedullary osteosynthesis is superior in the treatment of trochanteric fractures because of its lower complication rate, absence of differences in functional outcome compared to intramedullary osteosynthesis, and lower costs [4, 8, 9]. Extramedullary implants have been improved to provide more stability. Angular stable dynamic hip screw systems with angular stable trochanteric plates are intended to prevent secondary impaction, varisation and shortening of the fracture, as well as medialisation of the femoral shaft, especially in unstable fractures. Finally, there are several studies reporting almost no difference between both procedures [10–12]. From the biomechanical point of view, there are almost no studies comparing intramedullary and extramedullary osteosynthesis in trochanteric fracture treatment. In 1998, Go¨tze et al. investigated the failure load of different osteosyntheses of per- and subtrochanteric fractures [13]. They showed a much higher loadability of intramedullary implants (proximal femoral nail, gamma nail) compared to extramedullary implants (dynamic hip screw). However, today there are new angular stable implants, which might improve the extramedullary fixation. The aim of this in vitro biomechanical study was to compare extramedullary and intramedullary fixation of pertrochanteric fractures regarding stiffness, survival rates, failure loads, and failure mechanisms. We hypothesized that both procedures would provide comparable results except for a higher failure load of the intramedullary implant.
Materials and methods A total of twelve fresh-frozen human femurs (five female and seven male; mean age 64.4 years; SD 15.8; range 45–90) were included into the study. The specimens were removed postmortem at the local Institute of Forensic Medicine and stored at -20 °C. Since pertrochanteric fractures are often associated with osteoporosis, specimens were collected from donors with expected osteoporotic bone mass. Specimens with pre-existing fractures, bone deformities, or other evidence of impaired stability were excluded. All femurs were randomized into two groups based on the volumetric bone mineral density (vBMD), which was measured using quantitative computed tomography (qCT). These were either extramedullary or intramedullary osteosynthesis. Since trochanteric fractures are
123
often associated with osteoporosis and possible cutting out of the lag screw is significantly influenced by the bone quality, an osteoporotic bone model was chosen. After thawing the fresh-frozen specimens overnight at room temperature, the surrounding soft tissues were removed. The specimens were reduced by resecting the distal femoral condyles 30 cm below the proximal tip of the greater trochanter. Next, the distal 10 cm of the diaphysis was embedded in steel cylinders using polyurethane (Ureol FC 53, Go¨ßl & Pfaff, Karlskron, Germany). To imitate a single-leg stance with bending moments in sagittal and frontal planes and rotational moments under axial loading, the proximal femurs were tilted 10° laterally and 10° posteriorly with relation to the diaphyseal axis using a standardized positioning device [14, 15]. A dynamic hip screw (DHS) with an angular stable plate and helical blade (LCP 4 hole plate, 130°, Synthes GmbH, Umkirch, Germany) was used for extramedullary fixation. Since the DHS blade shows a higher rotational stability and resistance to push out compared to the conventional screw [16, 17], we decided to use the helical blade to achieve the most stable conditions for the extramedullary osteosynthesis. Furthermore, an additional, angular stable, trochanteric stabilization plate was implanted to ensure maximum stability. Implantation was performed following the surgical technique recommended by the manufacturer. After reaming for the helical blade and prior to implantation, a standardized rotationally unstable fracture (AO31-A2.3), as described in prior studies [13, 14], was created using an oscillating saw. The first osteotomy was made at an angle of 20° to the axis of the femur. It started from 1 cm below the greater trochanter went through the intertrochanteric ridge and ended at the bottom of the lesser trochanter. An additional superior and inferior wedge was removed by an osteotomy from the tip of the greater trochanter towards the middle of the diaphysis to the first osteotomy and from there to the top of the lesser trochanter (Fig. 1). Afterwards, the DHS system was implanted. An anatomical reduction was ensured due to reaming for the blade before generating the fracture. The Trigen Intertan nail (nail) with lag and compression screw (10 mm 9 18 cm, 125°, Smith & Nephew GmbH, Marl, Germany) was used for intramedullary fixation. The nail was implanted following the surgical technique recommended by the manufacturer. After reaming for the lag and compression screw the implant was removed and the fracture was created exactly in the same way as described for the DHS group. By reinserting the nail and implantation of the lag and compression screw, the fracture was anatomically restored and stabilized (Fig. 2). The blade of the DHS as well as the lag and compression screw of the nail was inserted center to center in the
Arch Orthop Trauma Surg
Fig. 1 Specimen with marked saw cuts, which were made to create an unstable pertrochanteric fracture (AO31-A2.3) Fig. 3 Biomechanical testing setup. The specimen is mounted in a servohydraulic testing machine. Loading is applied with a customized spherical shell
Fig. 2 Anteroposterior femur radiographs showing a specimen treated with the dynamic hip screw (a) and the Intertan nail (b)
femoral head. The tip–apex distance (TAD), as the sum of the distance from the tip of the blade/screw to the apex of the femoral head determined in anteroposterior and lateral radiographs and controlled for magnification, was measured in each case. For biomechanical testing, the specimens were mounted in a servohydraulic testing machine (MTS 858.2, Eden Prairie, MN, USA). Femurs were placed on an x–y table and load was applied using a custom-made spherical shell (Fig. 3). First, the native stiffness was measured for each femur by applying a non-destructive swelling axial load (50–700 N), for 10 cycles. To compare the native and postoperative status, stiffness was determined again following instrumentation. Afterwards, the constructs were loaded up to 1400 N for 10,000 cycles at 2 Hz, which should simulate the postoperative stress of a 70 kg person
during the first 4–6 weeks after fracture fixation [15]. Failure during cyclic testing was defined as an axial displacement of more than 20 mm. A load to failure test with a constant speed of 4.6 mm/s was performed in the surviving femurs. Radiographs in two planes were taken during implantation, after fixation and after cyclic testing or load to failure, respectively. Thus, implant position before and after testing, TAD, and failure mode could be determined. The specimens were kept moist with saline solution during preparation as well as during the whole testing procedure. Statistical analysis was performed using the software package IBM SPSS Statistics 21. Normal distribution was investigated using the Shapiro–Wilk test. Homogeneity of variances was tested by Levene’s test. The students t test was used to compare normally distributed results, whereas non-normally distributed results were compared with a Mann–Whitney U test. A regression analysis was conducted for correlation of BMD and native stiffness as well as postoperative stiffness and load to failure. An analysis of covariance was performed to determine the influence of the implant type on the postoperative stiffness controlling for native stiffness. The type I error probability was set to a = 0.05 for all tests. Approval of the ethics committee for this study was granted.
123
Arch Orthop Trauma Surg Table 1 Native stiffness, postoperative stiffness, and load to failure of the DHS and nail group Native stiffness (N/mm)
Postoperative stiffness (N/mm)
Tip–apex distance (mm)
Load to failure (N)
DHS
857.4 ± 288.4
395 ± 116.1
15.7 ± 1.6
2778.2 ± 196.8
Nail
907.3 ± 173.4
480.5 ± 139.6
15.5 ± 4.2
8480.8 ± 1238.9
Significant differences are marked in italics
Results The mean bone density in the DHS group (300.6 mg/cm3, SD 51.9; range 211.7–348.6) and the nail group (305.8 mg/ cm3, SD 50.2; range 220.3–344.8) were similar (p = 0.818). Mean age of the DHS group was 73.2 years (SD 14.6; range 47–90) and of the nail group 59.7 years (SD 9.1; range 45–69). The mean age of both groups differed significantly due to matching by bone density (p = 0.041). Each group contained one specimen with a bone density below 250 mg/cm3, which can be assumed as a high risk factor for cutting out [18]. TAD in the DHS group (15.7 mm, SD 1.6; range 13.5–18.3), and the nail group (15.5 mm, SD 4.2; range 11.4–23.3) were also similar (p = 0.923) (Table 1). Both groups were comparable with regard to their native stiffness (DHS 857.4 N/mm; SD 288.4; range 651.2–873.7 vs. nail 907.3 N/mm; SD 173.4; range 674.9–1084.8; p = 0.724). Instrumentation after fracture showed a decrease in stiffness for both groups (native 882.4 N/mm; SD 228.4; range 651.2–1416.4 vs. postoperative 437.8 N/ mm; SD 130.3; range 258.3–718; p \ 0.001; Fig. 4). No significance between the postoperative stiffness between the DHS and the nail group was observed (DHS 395 N/ mm; SD 116.1; range 258.3–466.8 vs. nail 480.5 N/mm; SD 139.6; range 297.9–718; p = 0.275) (Table 1), even if the native stiffness was used as covariate. One femur in the DHS group failed during cyclic loading after 2268 cycles due to cutting out of the helical blade. The bone density of this specimen was 211.7 mg/ cm3; the lowest value of all tested specimens. All other specimens from both, the DHS and nail groups, survived the 10,000 loading cycles. Load to failure for the specimens surviving the cyclic loading was lower in the DHS group (2778.2 N, SD 196.8; range 2608.8–3066) than in the nail group (8480.8 N, SD 1238.9; range 6912.8–9687; p = 0.008) (Table 1; Fig. 5). Failure in the DHS group occurred by plastic deformation of the blade in two cases and fracture of the shaft in the region of the locking screws in three cases. The main failure mechanism in the nail group was fracture of the femoral shaft beginning at the distal locking screw, which occurred in 5 cases. In one specimen, the lag screw bent during testing. There was no significant correlation between the load to failure and the bone density
123
Fig. 4 The postoperative stiffness of both constructs was significantly reduced (*p \ 0.001)
Fig. 5 Boxplot diagram showing the different failure loads for the dynamic hip screw and the Intertan nail (*p = 0.008)
Arch Orthop Trauma Surg
(r = -0.199, p = 0.581), the TAD (r = -0.172, p = 0.634), the native stiffness (r = 0.376, p = 0.284), or the postoperative stiffness (r = 0.181, p = 0.617). Furthermore, no influence of the bone density on the native stiffness was detected.
Discussion There are many different implants and operative procedures available to treat pertrochanteric fractures with common discussions regarding the optimal selection of choice. Current data show an increasing use of cephalomedullary nails despite a lack of evidence in the literature since several studies describe the dynamic hip screw to be superior [4, 5, 8]. Regardless of the implant, cutting out of the lag screw during postoperative mobilization is one of the leading failure mechanisms [19, 20]. In this study, a cyclic loading with an axial force up to 1400 N was performed for 10,000 cycles. Tilting of the specimens as described above simulated a one-leg stance. This loading condition was chosen to simulate full postoperative weight bearing of the hip during careful walking [21, 22]. The 10,000 cycles roughly represent the number of steps taken during fracture consolidation in the first 6 weeks after operation [23]. All specimens of the nail group survived this testing, while the sample with the lowest vBMD (211.7 mg/cm3) and the highest age (90 years) of the DHS group failed after 2268 cycles due to cutting out. Bonnaire et al. showed a clear correlation between BMD and cutting out and assumed values below 250 mg/cm3 as particularly critical [18]. One specimen of the nail group also met this criterion, but showed no cutting out during cyclic testing. Furthermore, the TAD and the Parker’s ratio as references for correct lag screw length and positioning in the femoral head are aspects influencing the cutting out risk [24]. In this study, all lag screws were inserted in center-tocenter position and the TAD was measured to be less than 24 mm in every specimen. Therefore, no influence of these factors on the results was observed. Compared to conventional lag screws, the helical blade, which was used in the DHS group, showed better results in many studies regarding cutting out, push out, and pull out [16, 17]. In bones with heavy osteoporosis, the helical blade with additional cement augmentation might represent the implant of choice [14, 25]. However, there is no study comparing the helical blade with a two-screw system as used for the Intertan nail. Since our study contained only two specimens with comparably poor bone quality and cutting out occurred in the worst specimen, we cannot draw any conclusion with regard to differences in cutting out between both tested groups.
The stiffness of all femurs was significantly reduced after osteosynthesis compared to the native status. This loss can be explained by the missing medial support due to the unstable AO31-A2.3 fracture, which was generated prior to testing. Therefore, the stiffness is essentially determined by the bending of the lag screw and the bending of the femoral shaft. No difference in postoperative stiffness between both groups was observed. This indicates that none of the two treatment procedures is superior with regard to the postoperative stability, which is essential for the initial fracture healing. The load to failure test showed significant differences between the DHS and the nail group. With a failure load of 8480.8 N, the nail group resists a more than three times higher force compared to the DHS group (2778.2 N). These results are similar to the ones of Go¨tze et al., who reported a significantly lower load to failure for DHS osteosynthesis compared with intramedullary osteosynthesis in a biomechanical in vitro testing with pertrochanteric fractures [13]. Shaft fractures beginning at the distal locking screw were the main reason for failure in the nail group. In the DHS group, shaft fractures in the area of the locking screws as well as bending of the helical blade led to failure of the construct. The significant difference of the failure load between extramedullary and intramedullary osteosynthesis can be explained by their divergent biomechanical principles. The shorter distance between the lag screw and the force-absorbing nail together with the advantage of intramedullary force transmission leads to the very high failure loads. As in this study, an intramedullary nail is usually longer than an extramedullary implant. This might result in a better distribution of the acting force and thus in a higher strength. Since pertrochanteric fractures particularly occur in the elderly, immediate stability, which allows full weight bearing, is essential. Bergmann et al. described a loading of the hip joint during walking of approximately 300 % of body weight [26]. Therefore, the osteosynthesis should at least withstand 2060 N, assuming a body weight of 70 kg. In case of stumbling, the forces can increase up to 700 % of the body weight [26], which would correspond to 4800 N in a 70 kg person. Considering the results of this study, the DHS would not withstand such strong forces. Clinical studies do not report this as a problem for the DHS and peri-implant fractures are more often described for cephalomedullary implants [11, 12]. This might be due to the fact that the effect of muscles was not considered in this study, which is one of the limitations. A biomechanical in vitro study is not able to completely simulate all in vivo conditions. Because of the use of fresh-frozen human specimens and the removal of soft tissues, the muscles and ligaments did not contribute to stability. Furthermore, the positioning of the implants was done prior to fracture creation, which might lead to
123
Arch Orthop Trauma Surg
different results in implant fixation strength than reduction and fixation after fracturing. The small sample size is dictated by the availability of human specimens and was not checked using an a priori power analysis. Furthermore, physiological mobilization might show different loading conditions as compared with simulated mobilization during axial loading in this study. In conclusion, the extramedullary and intramedullary osteosynthesis showed comparable results regarding the postoperative stiffness and survival rates during cyclic testing. The load to failure was significantly lower in the extramedullary group compared to the intramedullary group. Even so, this difference is not reflected as a problem in clinical practice, it might indicate a higher stability of intramedullary osteosynthesis in unstable pertrochanteric fractures. Acknowledgments acknowledged.
Funding from the state of Hamburg is kindly
Conflict of interest No potential conflict of interest is declared by all authors.
References 1. Kannus P, Parkkari J, Sievanen H, Heinonen A, Vuori I, Jarvinen M (1996) Epidemiology of hip fractures. Bone 18(1 Suppl):57S– 63S 2. White SM, Griffiths R (2011) Projected incidence of proximal femoral fracture in England: a report from the NHS hip fracture anaesthesia network (HIPFAN). Injury 42(11):1230–1233 3. Anglen JO, Weinstein JN (2008) Nail or plate fixation of intertrochanteric hip fractures: changing pattern of practice. A review of the American Board of Orthopaedic Surgery Database. J Bone Joint Surg Am 90(4):700–707 4. Parker MJ, Handoll HH (2010) Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 9:CD000093 5. Aktselis I, Kokoroghiannis C, Fragkomichalos E, Koundis G, Deligeorgis A, Daskalakis E et al (2014) Prospective randomised controlled trial of an intramedullary nail versus a sliding hip screw for intertrochanteric fractures of the femur. Int Orthop 38(1):155–161 6. Shen L, Zhang Y, Shen Y, Cui Z (2013) Antirotation proximal femoral nail versus dynamic hip screw for intertrochanteric fractures: a meta-analysis of randomized controlled studies. Orthop Traumatol Surg Res 99(4):377–383 7. Mereddy P, Kamath S, Ramakrishnan M, Malik H, Donnachie N (2009) The AO/ASIF proximal femoral nail antirotation (PFNA): a new design for the treatment of unstable proximal femoral fractures. Injury 40(4):428–432 8. Barton TM, Gleeson R, Topliss C, Greenwood R, Harries WJ, Chesser TJ (2010) A comparison of the long gamma nail with the sliding hip screw for the treatment of AO/OTA 31-A2 fractures of the proximal part of the femur: a prospective randomized trial. J Bone Joint Surg Am 92(4):792–798 9. Yli-Kyyny TT, Sund R, Juntunen M, Salo JJ, Kroger HP (2012) Extra- and intramedullary implants for the treatment of pertrochanteric fractures—results from a Finnish National Database Study of 14,915 patients. Injury 43(12):2156–2160
123
10. Huang X, Leung F, Xiang Z, Tan PY, Yang J, Wei DQ et al (2013) Proximal femoral nail versus dynamic hip screw fixation for trochanteric fractures: a meta-analysis of randomized controlled trials. Sci World J 2013:805805 11. Liu M, Yang Z, Pei F, Huang F, Chen S, Xiang Z (2010) A metaanalysis of the gamma nail and dynamic hip screw in treating peritrochanteric fractures. Int Orthop 34(3):323–328 12. Matre K, Vinje T, Havelin LI, Gjertsen JE, Furnes O, Espehaug B et al (2013) TRIGEN INTERTAN intramedullary nail versus sliding hip screw: a prospective, randomized multicenter study on pain, function, and complications in 684 patients with an intertrochanteric or subtrochanteric fracture and one year of followup. J Bone Joint Surg Am 95(3):200–208 13. Go¨tze B, Bonnaire F, Weise K, Friedl HP (1998) Loadability of osteosynthesis of unstable per- and subtrochanteric fractures: an experimental study testing the proximal femoral nail (PFN), the gamma-nail, the DHS/trochanteric stabilization plate, the 95°angled blade plate and the UFN/spiral blade. Akt Traumatol 28:197–204 14. Fensky F, Nuchtern JV, Kolb JP, Huber S, Rupprecht M, Jauch SY et al (2013) Cement augmentation of the proximal femoral nail antirotation for the treatment of osteoporotic pertrochanteric fractures—a biomechanical cadaver study. Injury 44(6):802–807 15. Rupprecht M, Grossterlinden L, Ruecker AH, de Oliveira AN, Sellenschloh K, Nuchtern J et al (2011) A comparative biomechanical analysis of fixation devices for unstable femoral neck fractures: the Intertan versus cannulated screws or a dynamic hip screw. J Trauma 71(3):625–634 16. Luo Q, Yuen G, Lau TW, Yeung K, Leung F (2013) A biomechanical study comparing helical blade with screw design for sliding hip fixations of unstable intertrochanteric fractures. Sci World J 2013:351936 17. O’Neill F, Condon F, McGloughlin T, Lenehan B, Coffey JC, Walsh M (2011) Dynamic hip screw versus DHS blade: a biomechanical comparison of the fixation achieved by each implant in bone. J Bone Joint Surg Br 93(5):616–621 18. Bonnaire F, Weber A, Bosl O, Eckhardt C, Schwieger K, Linke B (2007) ‘‘Cutting out’’ in pertrochanteric fractures—problem of osteoporosis? Unfallchirurg 110(5):425–432 19. Haynes RC, Poll RG, Miles AW, Weston RB (1997) Failure of femoral head fixation: a cadaveric analysis of lag screw cut-out with the gamma locking nail and AO dynamic hip screw. Injury 28(5–6):337–341 20. Pervez H, Parker MJ, Vowler S (2004) Prediction of fixation failure after sliding hip screw fixation. Injury 35(10):994–998 21. Bergmann G, Deuretzbacher G, Heller M, Graichen F, Rohlmann A, Strauss J et al (2001) Hip contact forces and gait patterns from routine activities. J Biomech 34(7):859–871 22. Duda GN, Schneider E, Chao EY (1997) Internal forces and moments in the femur during walking. J Biomech 30(9):933–941 23. Kubiak EN, Bong M, Park SS, Kummer F, Egol K, Koval KJ (2004) Intramedullary fixation of unstable intertrochanteric hip fractures: one or two lag screws. J Orthop Trauma 18(1):12–17 24. Lobo-Escolar A, Joven E, Iglesias D, Herrera A (2010) Predictive factors for cutting-out in femoral intramedullary nailing. Injury 41(12):1312–1316 25. von der LP, Gisep A, Boner V, Windolf M, Appelt A, Suhm N (2006) Biomechanical evaluation of a new augmentation method for enhanced screw fixation in osteoporotic proximal femoral fractures. J Orthop Res 24(12):2230–2237 26. Bergmann G, Graichen F, Rohlmann A (1993) Hip joint loading during walking and running, measured in two patients. J Biomech 26(8):969–990
TRAUMA SURGERY
Extra- vs. intramedullary treatment of pertrochanteric fractures: a biomechanical in vitro study comparing dynamic hip screw and intramedullary nail Lukas Weiser1 • Andreas A. Ruppel2 • Jakob V. Nu¨chtern1 • Kay Sellenschloh3 • Johannes Zeichen2 • Klaus Pu¨schel4 • Michael M. Morlock3 • Wolfgang Lehmann1
Received: 15 February 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Introduction Due to the demographic trend, pertrochanteric fractures of the femur will gain increasing importance in the future. Both extra- and intramedullary implants are used with good results in the treatment of these fractures. New, angular stable extramedullary implants promise increased postoperative stability even with unstable fractures. Additional trochanteric plates are intended to prevent secondary impaction, varisation and shortening of the fracture, as well as medialisation of the femoral shaft. The aim of this study was to perform a biomechanical comparison of both procedures regarding their postoperative stability and failure mechanisms. Materials and methods Twelve fresh-frozen human femurs were randomized into two groups based on the volumetric bone mineral density (vBMD). Standardized pertrochanteric fractures (AO31-A2.3) were generated and treated either with an angular stable dynamic hip screw (DHS) or an intramedullary nail (nail). Correct implant position and the tip–apex distance (TAD) were controlled postoperatively using X-ray. Specimens were mounted in a
servohydraulic testing machine and an axial loading was applied according to a single-leg stance model. Both groups were biomechanically compared with regard to native and postoperative stiffness, survival during cyclic testing, load to failure, and failure mechanisms. Results TAD, vBMD, and native stiffness were similar for both groups. The stiffness decreased significantly from native to postoperative state in all specimens (p \ 0.001). The postoperative stiffness of both groups varied non-significantly (p = 0.275). The failure loads for specimens treated with the nail were significantly higher than for those treated with the DHS (8480.8 ± 1238.9 N vs. 2778.2 ± 196.8 N; p = 0.008). Conclusions Extra- and intramedullary osteosynthesis showed comparable results as regards postoperative stiffness and survival during cyclic testing. Since the failure load of the nail was significantly higher in the tested AO31A2.3 fracture model, we conclude that intramedullary implants should be preferred in these, unstable, fractures.
L. Weiser and A. A. Ruppel contributed equally and therefore share first authorship.
Keywords Pertrochanteric fracture Hip fracture Sliding hip screw Dynamic hip screw Intramedullary nail Intertan nail
& Lukas Weiser [email protected]
Introduction
1
Department of Trauma-, Hand- and Reconstructive Surgery, University Medical Center Hamburg-Eppendorf, Martinistraße 52, 20246 Hamburg, Germany
2
Trauma and Orthopedic Department, Johannes Wesling Klinikum Minden, Minden, Germany
3
Institute of Biomechanics, TUHH Hamburg University of Technology, Hamburg, Germany
4
Institute of Forensic Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Pertrochanteric fractures are one of the most common injuries associated with osteoporosis. Due to the increasing population older than 65 years, these fractures will gain clinical as well as economic importance in the future [1, 2]. Both intramedullary and extramedullary implants are available for the treatment of trochanteric fractures; however, there is no gold standard. Recent studies show that the use of intramedullary nails has overtaken the use of
123
Arch Orthop Trauma Surg
extramedullary implants [3, 4]. Advantages such as better reconstituting of patients to their pre-operative state, lesser blood loss, and fewer complications have been described [5, 6]. Furthermore, cephalomedullary nails may be beneficial to the treatment of unstable and subtrochanteric fractures [4, 7]. On the contrary, it has been shown that extramedullary osteosynthesis is superior in the treatment of trochanteric fractures because of its lower complication rate, absence of differences in functional outcome compared to intramedullary osteosynthesis, and lower costs [4, 8, 9]. Extramedullary implants have been improved to provide more stability. Angular stable dynamic hip screw systems with angular stable trochanteric plates are intended to prevent secondary impaction, varisation and shortening of the fracture, as well as medialisation of the femoral shaft, especially in unstable fractures. Finally, there are several studies reporting almost no difference between both procedures [10–12]. From the biomechanical point of view, there are almost no studies comparing intramedullary and extramedullary osteosynthesis in trochanteric fracture treatment. In 1998, Go¨tze et al. investigated the failure load of different osteosyntheses of per- and subtrochanteric fractures [13]. They showed a much higher loadability of intramedullary implants (proximal femoral nail, gamma nail) compared to extramedullary implants (dynamic hip screw). However, today there are new angular stable implants, which might improve the extramedullary fixation. The aim of this in vitro biomechanical study was to compare extramedullary and intramedullary fixation of pertrochanteric fractures regarding stiffness, survival rates, failure loads, and failure mechanisms. We hypothesized that both procedures would provide comparable results except for a higher failure load of the intramedullary implant.
Materials and methods A total of twelve fresh-frozen human femurs (five female and seven male; mean age 64.4 years; SD 15.8; range 45–90) were included into the study. The specimens were removed postmortem at the local Institute of Forensic Medicine and stored at -20 °C. Since pertrochanteric fractures are often associated with osteoporosis, specimens were collected from donors with expected osteoporotic bone mass. Specimens with pre-existing fractures, bone deformities, or other evidence of impaired stability were excluded. All femurs were randomized into two groups based on the volumetric bone mineral density (vBMD), which was measured using quantitative computed tomography (qCT). These were either extramedullary or intramedullary osteosynthesis. Since trochanteric fractures are
123
often associated with osteoporosis and possible cutting out of the lag screw is significantly influenced by the bone quality, an osteoporotic bone model was chosen. After thawing the fresh-frozen specimens overnight at room temperature, the surrounding soft tissues were removed. The specimens were reduced by resecting the distal femoral condyles 30 cm below the proximal tip of the greater trochanter. Next, the distal 10 cm of the diaphysis was embedded in steel cylinders using polyurethane (Ureol FC 53, Go¨ßl & Pfaff, Karlskron, Germany). To imitate a single-leg stance with bending moments in sagittal and frontal planes and rotational moments under axial loading, the proximal femurs were tilted 10° laterally and 10° posteriorly with relation to the diaphyseal axis using a standardized positioning device [14, 15]. A dynamic hip screw (DHS) with an angular stable plate and helical blade (LCP 4 hole plate, 130°, Synthes GmbH, Umkirch, Germany) was used for extramedullary fixation. Since the DHS blade shows a higher rotational stability and resistance to push out compared to the conventional screw [16, 17], we decided to use the helical blade to achieve the most stable conditions for the extramedullary osteosynthesis. Furthermore, an additional, angular stable, trochanteric stabilization plate was implanted to ensure maximum stability. Implantation was performed following the surgical technique recommended by the manufacturer. After reaming for the helical blade and prior to implantation, a standardized rotationally unstable fracture (AO31-A2.3), as described in prior studies [13, 14], was created using an oscillating saw. The first osteotomy was made at an angle of 20° to the axis of the femur. It started from 1 cm below the greater trochanter went through the intertrochanteric ridge and ended at the bottom of the lesser trochanter. An additional superior and inferior wedge was removed by an osteotomy from the tip of the greater trochanter towards the middle of the diaphysis to the first osteotomy and from there to the top of the lesser trochanter (Fig. 1). Afterwards, the DHS system was implanted. An anatomical reduction was ensured due to reaming for the blade before generating the fracture. The Trigen Intertan nail (nail) with lag and compression screw (10 mm 9 18 cm, 125°, Smith & Nephew GmbH, Marl, Germany) was used for intramedullary fixation. The nail was implanted following the surgical technique recommended by the manufacturer. After reaming for the lag and compression screw the implant was removed and the fracture was created exactly in the same way as described for the DHS group. By reinserting the nail and implantation of the lag and compression screw, the fracture was anatomically restored and stabilized (Fig. 2). The blade of the DHS as well as the lag and compression screw of the nail was inserted center to center in the
Arch Orthop Trauma Surg
Fig. 1 Specimen with marked saw cuts, which were made to create an unstable pertrochanteric fracture (AO31-A2.3) Fig. 3 Biomechanical testing setup. The specimen is mounted in a servohydraulic testing machine. Loading is applied with a customized spherical shell
Fig. 2 Anteroposterior femur radiographs showing a specimen treated with the dynamic hip screw (a) and the Intertan nail (b)
femoral head. The tip–apex distance (TAD), as the sum of the distance from the tip of the blade/screw to the apex of the femoral head determined in anteroposterior and lateral radiographs and controlled for magnification, was measured in each case. For biomechanical testing, the specimens were mounted in a servohydraulic testing machine (MTS 858.2, Eden Prairie, MN, USA). Femurs were placed on an x–y table and load was applied using a custom-made spherical shell (Fig. 3). First, the native stiffness was measured for each femur by applying a non-destructive swelling axial load (50–700 N), for 10 cycles. To compare the native and postoperative status, stiffness was determined again following instrumentation. Afterwards, the constructs were loaded up to 1400 N for 10,000 cycles at 2 Hz, which should simulate the postoperative stress of a 70 kg person
during the first 4–6 weeks after fracture fixation [15]. Failure during cyclic testing was defined as an axial displacement of more than 20 mm. A load to failure test with a constant speed of 4.6 mm/s was performed in the surviving femurs. Radiographs in two planes were taken during implantation, after fixation and after cyclic testing or load to failure, respectively. Thus, implant position before and after testing, TAD, and failure mode could be determined. The specimens were kept moist with saline solution during preparation as well as during the whole testing procedure. Statistical analysis was performed using the software package IBM SPSS Statistics 21. Normal distribution was investigated using the Shapiro–Wilk test. Homogeneity of variances was tested by Levene’s test. The students t test was used to compare normally distributed results, whereas non-normally distributed results were compared with a Mann–Whitney U test. A regression analysis was conducted for correlation of BMD and native stiffness as well as postoperative stiffness and load to failure. An analysis of covariance was performed to determine the influence of the implant type on the postoperative stiffness controlling for native stiffness. The type I error probability was set to a = 0.05 for all tests. Approval of the ethics committee for this study was granted.
123
Arch Orthop Trauma Surg Table 1 Native stiffness, postoperative stiffness, and load to failure of the DHS and nail group Native stiffness (N/mm)
Postoperative stiffness (N/mm)
Tip–apex distance (mm)
Load to failure (N)
DHS
857.4 ± 288.4
395 ± 116.1
15.7 ± 1.6
2778.2 ± 196.8
Nail
907.3 ± 173.4
480.5 ± 139.6
15.5 ± 4.2
8480.8 ± 1238.9
Significant differences are marked in italics
Results The mean bone density in the DHS group (300.6 mg/cm3, SD 51.9; range 211.7–348.6) and the nail group (305.8 mg/ cm3, SD 50.2; range 220.3–344.8) were similar (p = 0.818). Mean age of the DHS group was 73.2 years (SD 14.6; range 47–90) and of the nail group 59.7 years (SD 9.1; range 45–69). The mean age of both groups differed significantly due to matching by bone density (p = 0.041). Each group contained one specimen with a bone density below 250 mg/cm3, which can be assumed as a high risk factor for cutting out [18]. TAD in the DHS group (15.7 mm, SD 1.6; range 13.5–18.3), and the nail group (15.5 mm, SD 4.2; range 11.4–23.3) were also similar (p = 0.923) (Table 1). Both groups were comparable with regard to their native stiffness (DHS 857.4 N/mm; SD 288.4; range 651.2–873.7 vs. nail 907.3 N/mm; SD 173.4; range 674.9–1084.8; p = 0.724). Instrumentation after fracture showed a decrease in stiffness for both groups (native 882.4 N/mm; SD 228.4; range 651.2–1416.4 vs. postoperative 437.8 N/ mm; SD 130.3; range 258.3–718; p \ 0.001; Fig. 4). No significance between the postoperative stiffness between the DHS and the nail group was observed (DHS 395 N/ mm; SD 116.1; range 258.3–466.8 vs. nail 480.5 N/mm; SD 139.6; range 297.9–718; p = 0.275) (Table 1), even if the native stiffness was used as covariate. One femur in the DHS group failed during cyclic loading after 2268 cycles due to cutting out of the helical blade. The bone density of this specimen was 211.7 mg/ cm3; the lowest value of all tested specimens. All other specimens from both, the DHS and nail groups, survived the 10,000 loading cycles. Load to failure for the specimens surviving the cyclic loading was lower in the DHS group (2778.2 N, SD 196.8; range 2608.8–3066) than in the nail group (8480.8 N, SD 1238.9; range 6912.8–9687; p = 0.008) (Table 1; Fig. 5). Failure in the DHS group occurred by plastic deformation of the blade in two cases and fracture of the shaft in the region of the locking screws in three cases. The main failure mechanism in the nail group was fracture of the femoral shaft beginning at the distal locking screw, which occurred in 5 cases. In one specimen, the lag screw bent during testing. There was no significant correlation between the load to failure and the bone density
123
Fig. 4 The postoperative stiffness of both constructs was significantly reduced (*p \ 0.001)
Fig. 5 Boxplot diagram showing the different failure loads for the dynamic hip screw and the Intertan nail (*p = 0.008)
Arch Orthop Trauma Surg
(r = -0.199, p = 0.581), the TAD (r = -0.172, p = 0.634), the native stiffness (r = 0.376, p = 0.284), or the postoperative stiffness (r = 0.181, p = 0.617). Furthermore, no influence of the bone density on the native stiffness was detected.
Discussion There are many different implants and operative procedures available to treat pertrochanteric fractures with common discussions regarding the optimal selection of choice. Current data show an increasing use of cephalomedullary nails despite a lack of evidence in the literature since several studies describe the dynamic hip screw to be superior [4, 5, 8]. Regardless of the implant, cutting out of the lag screw during postoperative mobilization is one of the leading failure mechanisms [19, 20]. In this study, a cyclic loading with an axial force up to 1400 N was performed for 10,000 cycles. Tilting of the specimens as described above simulated a one-leg stance. This loading condition was chosen to simulate full postoperative weight bearing of the hip during careful walking [21, 22]. The 10,000 cycles roughly represent the number of steps taken during fracture consolidation in the first 6 weeks after operation [23]. All specimens of the nail group survived this testing, while the sample with the lowest vBMD (211.7 mg/cm3) and the highest age (90 years) of the DHS group failed after 2268 cycles due to cutting out. Bonnaire et al. showed a clear correlation between BMD and cutting out and assumed values below 250 mg/cm3 as particularly critical [18]. One specimen of the nail group also met this criterion, but showed no cutting out during cyclic testing. Furthermore, the TAD and the Parker’s ratio as references for correct lag screw length and positioning in the femoral head are aspects influencing the cutting out risk [24]. In this study, all lag screws were inserted in center-tocenter position and the TAD was measured to be less than 24 mm in every specimen. Therefore, no influence of these factors on the results was observed. Compared to conventional lag screws, the helical blade, which was used in the DHS group, showed better results in many studies regarding cutting out, push out, and pull out [16, 17]. In bones with heavy osteoporosis, the helical blade with additional cement augmentation might represent the implant of choice [14, 25]. However, there is no study comparing the helical blade with a two-screw system as used for the Intertan nail. Since our study contained only two specimens with comparably poor bone quality and cutting out occurred in the worst specimen, we cannot draw any conclusion with regard to differences in cutting out between both tested groups.
The stiffness of all femurs was significantly reduced after osteosynthesis compared to the native status. This loss can be explained by the missing medial support due to the unstable AO31-A2.3 fracture, which was generated prior to testing. Therefore, the stiffness is essentially determined by the bending of the lag screw and the bending of the femoral shaft. No difference in postoperative stiffness between both groups was observed. This indicates that none of the two treatment procedures is superior with regard to the postoperative stability, which is essential for the initial fracture healing. The load to failure test showed significant differences between the DHS and the nail group. With a failure load of 8480.8 N, the nail group resists a more than three times higher force compared to the DHS group (2778.2 N). These results are similar to the ones of Go¨tze et al., who reported a significantly lower load to failure for DHS osteosynthesis compared with intramedullary osteosynthesis in a biomechanical in vitro testing with pertrochanteric fractures [13]. Shaft fractures beginning at the distal locking screw were the main reason for failure in the nail group. In the DHS group, shaft fractures in the area of the locking screws as well as bending of the helical blade led to failure of the construct. The significant difference of the failure load between extramedullary and intramedullary osteosynthesis can be explained by their divergent biomechanical principles. The shorter distance between the lag screw and the force-absorbing nail together with the advantage of intramedullary force transmission leads to the very high failure loads. As in this study, an intramedullary nail is usually longer than an extramedullary implant. This might result in a better distribution of the acting force and thus in a higher strength. Since pertrochanteric fractures particularly occur in the elderly, immediate stability, which allows full weight bearing, is essential. Bergmann et al. described a loading of the hip joint during walking of approximately 300 % of body weight [26]. Therefore, the osteosynthesis should at least withstand 2060 N, assuming a body weight of 70 kg. In case of stumbling, the forces can increase up to 700 % of the body weight [26], which would correspond to 4800 N in a 70 kg person. Considering the results of this study, the DHS would not withstand such strong forces. Clinical studies do not report this as a problem for the DHS and peri-implant fractures are more often described for cephalomedullary implants [11, 12]. This might be due to the fact that the effect of muscles was not considered in this study, which is one of the limitations. A biomechanical in vitro study is not able to completely simulate all in vivo conditions. Because of the use of fresh-frozen human specimens and the removal of soft tissues, the muscles and ligaments did not contribute to stability. Furthermore, the positioning of the implants was done prior to fracture creation, which might lead to
123
Arch Orthop Trauma Surg
different results in implant fixation strength than reduction and fixation after fracturing. The small sample size is dictated by the availability of human specimens and was not checked using an a priori power analysis. Furthermore, physiological mobilization might show different loading conditions as compared with simulated mobilization during axial loading in this study. In conclusion, the extramedullary and intramedullary osteosynthesis showed comparable results regarding the postoperative stiffness and survival rates during cyclic testing. The load to failure was significantly lower in the extramedullary group compared to the intramedullary group. Even so, this difference is not reflected as a problem in clinical practice, it might indicate a higher stability of intramedullary osteosynthesis in unstable pertrochanteric fractures. Acknowledgments acknowledged.
Funding from the state of Hamburg is kindly
Conflict of interest No potential conflict of interest is declared by all authors.
References 1. Kannus P, Parkkari J, Sievanen H, Heinonen A, Vuori I, Jarvinen M (1996) Epidemiology of hip fractures. Bone 18(1 Suppl):57S– 63S 2. White SM, Griffiths R (2011) Projected incidence of proximal femoral fracture in England: a report from the NHS hip fracture anaesthesia network (HIPFAN). Injury 42(11):1230–1233 3. Anglen JO, Weinstein JN (2008) Nail or plate fixation of intertrochanteric hip fractures: changing pattern of practice. A review of the American Board of Orthopaedic Surgery Database. J Bone Joint Surg Am 90(4):700–707 4. Parker MJ, Handoll HH (2010) Gamma and other cephalocondylic intramedullary nails versus extramedullary implants for extracapsular hip fractures in adults. Cochrane Database Syst Rev 9:CD000093 5. Aktselis I, Kokoroghiannis C, Fragkomichalos E, Koundis G, Deligeorgis A, Daskalakis E et al (2014) Prospective randomised controlled trial of an intramedullary nail versus a sliding hip screw for intertrochanteric fractures of the femur. Int Orthop 38(1):155–161 6. Shen L, Zhang Y, Shen Y, Cui Z (2013) Antirotation proximal femoral nail versus dynamic hip screw for intertrochanteric fractures: a meta-analysis of randomized controlled studies. Orthop Traumatol Surg Res 99(4):377–383 7. Mereddy P, Kamath S, Ramakrishnan M, Malik H, Donnachie N (2009) The AO/ASIF proximal femoral nail antirotation (PFNA): a new design for the treatment of unstable proximal femoral fractures. Injury 40(4):428–432 8. Barton TM, Gleeson R, Topliss C, Greenwood R, Harries WJ, Chesser TJ (2010) A comparison of the long gamma nail with the sliding hip screw for the treatment of AO/OTA 31-A2 fractures of the proximal part of the femur: a prospective randomized trial. J Bone Joint Surg Am 92(4):792–798 9. Yli-Kyyny TT, Sund R, Juntunen M, Salo JJ, Kroger HP (2012) Extra- and intramedullary implants for the treatment of pertrochanteric fractures—results from a Finnish National Database Study of 14,915 patients. Injury 43(12):2156–2160
123
10. Huang X, Leung F, Xiang Z, Tan PY, Yang J, Wei DQ et al (2013) Proximal femoral nail versus dynamic hip screw fixation for trochanteric fractures: a meta-analysis of randomized controlled trials. Sci World J 2013:805805 11. Liu M, Yang Z, Pei F, Huang F, Chen S, Xiang Z (2010) A metaanalysis of the gamma nail and dynamic hip screw in treating peritrochanteric fractures. Int Orthop 34(3):323–328 12. Matre K, Vinje T, Havelin LI, Gjertsen JE, Furnes O, Espehaug B et al (2013) TRIGEN INTERTAN intramedullary nail versus sliding hip screw: a prospective, randomized multicenter study on pain, function, and complications in 684 patients with an intertrochanteric or subtrochanteric fracture and one year of followup. J Bone Joint Surg Am 95(3):200–208 13. Go¨tze B, Bonnaire F, Weise K, Friedl HP (1998) Loadability of osteosynthesis of unstable per- and subtrochanteric fractures: an experimental study testing the proximal femoral nail (PFN), the gamma-nail, the DHS/trochanteric stabilization plate, the 95°angled blade plate and the UFN/spiral blade. Akt Traumatol 28:197–204 14. Fensky F, Nuchtern JV, Kolb JP, Huber S, Rupprecht M, Jauch SY et al (2013) Cement augmentation of the proximal femoral nail antirotation for the treatment of osteoporotic pertrochanteric fractures—a biomechanical cadaver study. Injury 44(6):802–807 15. Rupprecht M, Grossterlinden L, Ruecker AH, de Oliveira AN, Sellenschloh K, Nuchtern J et al (2011) A comparative biomechanical analysis of fixation devices for unstable femoral neck fractures: the Intertan versus cannulated screws or a dynamic hip screw. J Trauma 71(3):625–634 16. Luo Q, Yuen G, Lau TW, Yeung K, Leung F (2013) A biomechanical study comparing helical blade with screw design for sliding hip fixations of unstable intertrochanteric fractures. Sci World J 2013:351936 17. O’Neill F, Condon F, McGloughlin T, Lenehan B, Coffey JC, Walsh M (2011) Dynamic hip screw versus DHS blade: a biomechanical comparison of the fixation achieved by each implant in bone. J Bone Joint Surg Br 93(5):616–621 18. Bonnaire F, Weber A, Bosl O, Eckhardt C, Schwieger K, Linke B (2007) ‘‘Cutting out’’ in pertrochanteric fractures—problem of osteoporosis? Unfallchirurg 110(5):425–432 19. Haynes RC, Poll RG, Miles AW, Weston RB (1997) Failure of femoral head fixation: a cadaveric analysis of lag screw cut-out with the gamma locking nail and AO dynamic hip screw. Injury 28(5–6):337–341 20. Pervez H, Parker MJ, Vowler S (2004) Prediction of fixation failure after sliding hip screw fixation. Injury 35(10):994–998 21. Bergmann G, Deuretzbacher G, Heller M, Graichen F, Rohlmann A, Strauss J et al (2001) Hip contact forces and gait patterns from routine activities. J Biomech 34(7):859–871 22. Duda GN, Schneider E, Chao EY (1997) Internal forces and moments in the femur during walking. J Biomech 30(9):933–941 23. Kubiak EN, Bong M, Park SS, Kummer F, Egol K, Koval KJ (2004) Intramedullary fixation of unstable intertrochanteric hip fractures: one or two lag screws. J Orthop Trauma 18(1):12–17 24. Lobo-Escolar A, Joven E, Iglesias D, Herrera A (2010) Predictive factors for cutting-out in femoral intramedullary nailing. Injury 41(12):1312–1316 25. von der LP, Gisep A, Boner V, Windolf M, Appelt A, Suhm N (2006) Biomechanical evaluation of a new augmentation method for enhanced screw fixation in osteoporotic proximal femoral fractures. J Orthop Res 24(12):2230–2237 26. Bergmann G, Graichen F, Rohlmann A (1993) Hip joint loading during walking and running, measured in two patients. J Biomech 26(8):969–990
