Cement Augmentation of Implants—No General Cure in Osteoporotic Fracture Treatment. A Biomechanical Study on Non-Displaced Femoral Neck Fractures Ladina Hofmann-Fliri,1 Tomas I. Nicolino,1,2 Jorge Barla,2 Boyko Gueorguiev,1 R. Geoff Richards,1 Michael Blauth,3 Markus Windolf1 1
AO Research Institute Davos, Switzerland, 2Hospital Italiano de Buenos Aires, Argentina, 3Medical University of Innsbruck, Austria
Received 30 April 2015; accepted 24 June 2015 Published online 29 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22978
ABSTRACT: Femoral neck fractures in the elderly are a common problem in orthopedics. Augmentation of screw fixation with bone cement can provide better stability of implants and lower the risk of secondary displacement. This study aimed to investigate whether cement augmentation of three cannulated screws in non-displaced femoral neck fractures could increase implant fixation. A femoral neck fracture was simulated in six paired human cadaveric femora and stabilized with three 7.3 mm cannulated screws. Pairs were divided into two groups: conventional instrumentation versus additional cement augmentation of screw tips with 2 ml TraumacemVþ each. Biomechanical testing was performed by applying cyclic axial load until failure. Failure cycles, axial head displacement, screw angle changes, telescoping and screw cut-out were evaluated. Failure (15 mm actuator displacement) occurred in the augmented group at 12,500 cycles ( 2,480) compared to 15,625 cycles ( 4,215) in the non-augmented group (p ¼ 0.041). When comparing 3 mm vertical displacement of the head no significant difference (p ¼ 0.72) was detected between the survival curves of the two groups. At 8,500 loadcycles (early onset failure) the augmented group demonstrated a change in screw angle of 2.85˚ ( 0.84) compared to 1.15˚ ( 0.93) in the non-augmented group (p ¼ 0.013). The results showed no biomechanical advantage with respect to secondary displacement following augmentation of three cannulated screws in a non-displaced femoral neck fracture. Consequently, the indication for cement augmentation to enhance implant anchorage in osteoporotic bone has to be considered carefully taking into account fracture type, implant selection and biomechanical surrounding. ß 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 34:314–319, 2016. Keywords: femoral neck fracture; implant augmentation; cannulated screw; osteoporosis; PMMA
The annual worldwide incidence of hip fractures exceeds 1.7 million.1 Femoral neck fractures are one of the most frequent, accounting for 60% of all hip fractures and they are becoming more common as the proportion of elderly people increases.2–4 Their association with osteoporotic bone and the risks inherent to the injury and treatment renders their management difficult. One major concern with osteoporotic fracture fixation is related to obtaining secure anchorage of the implant due to the decrease in cortical thickness, cancellous bone density and increased porosity.5,6 The most commonly used implants to fix neck fractures are sliding hip screws and percutaneous cancellous screws. The use of cancellous screws can be a less expensive and less invasive option for non-displaced or minimally displaced fractures, particularly in comparison with total hip arthroplasty.7 However, complications such as secondary displacement are reported in 10–25% and non-union in 7–10% of the cases.3,7,8 Failures after internal fixation usually lead to revision surgery and total hip arthroplasty.3 Bone cement augmentation, as described for different applications in osteoporotic fracture fixation, can
Ladina Hofmann-Fliri and Tomas I. Nicolino contributed equally to this work. Conflict of interest: The authors are not compensated and there are no other institutional subsidies, corporate affiliations, or funding sources supporting this work unless clearly documented and disclosed. Grant sponsor: AOTRAUMA Network; Grant number: AR2008_01. Correspondence to: Markus Windolf (T: þ41-81-414-23-29; F: þ41-81-414-22-88; E-mail: [email protected]) # 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
314
JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
provide better implant stability at reduced risk for secondary displacement. Sermon et al.9 compared polymethylmethacrylate (PMMA) augmented helical blades of a proximal femoral nail with non-augmented blades in human cadaveric femoral heads and found 51% increased load cycles to failure due to application of small amounts of bone cement (3 ml) in the femoral head region. The effect was particularly accentuated in osteoporotic bone. Clinically, it could be shown in a prospective multicenter trial that the same procedure led to good functional results with no complications related to the cement augmentation.10 Other biomechanical studies report similar advantages and, hence, draw a consistent picture of the concept. This includes PMMA augmentation of a sliding hip screw,11,12 of proximal humerus plate screws,13 of distal femur and proximal tibia plate screws,14,15 of lag screws16 and of a tibiotalocalcaneal arthrodesis nail.17 Thus, there is reason to presume that augmenting the tips of cannulated cancellous screws for fixation of femoral neck fractures might improve implant anchorage in a similar manner. The aim of this study was therefore to experimentally evaluate the biomechanical potential of cement augmentation of three cannulated cancellous screws in the femoral head in a non-displaced proximal femoral neck fracture model (AO type 31 B2).
MATERIALS AND METHODS Six pairs of fresh frozen (20˚) human cadaveric femora without soft tissues were provided by the Institute of Pathology from the University Hospital Basel, Switzerland. The mean age was 81 years (minimum 76; maximum 90);
CEMENT AUGMENTATION OF IMPLANTS
three specimen pairs were from female donors and the other three from male donors. Bone mineral density (BMD) was measured in the femoral head for each specimen by peripheral quantitative computed tomography (pQCT) using an XtremeCT (SCANCO Medical AG, Br€ uttisellen, Switzerland). A femoral neck fracture was simulated by an osteotomy in all the specimens. The fracture was created at an angle of 45˚ to the horizontal plane.18 This corresponds to a fracture perpendicular to the axis of the femoral neck if a caputcollum-diaphyseal (CCD) angle of 135˚ is considered.19 A line which was drawn in the middle of the femoral neck, midway between the femoral head cartilage and the intertrochanteric line helped to create the osteotomy (Fig. 1). The fracture was created with an oscillating saw and was stabilized with a triangular configuration of three 7.3 mm cannulated titanium alloy cancellous hip screws (DepuySynthes Inc., Oberdorf, Switzerland) using parallel guide wires. An apex single screw was placed superiorly.20 One from each pair of femora was assigned into one of the two groups: augmented versus conventional screws. In the augmentation group, the cannulated screws were perforated with 4 holes of 2 mm diameter, starting 10 mm distal to the tip of the screw over a distance of 10 mm, alternating by 120˚ angles (Fig. 2). The screws were inserted into the femoral heads up to a distance of 5 mm from the articular surface.21,22 Augmentation was performed through a custom-made side-opening cannula of 2.5 mm diameter. Acrylic bone cement (Traumacem Vþ, DepuySynthes Inc., Oberdorf, Switzerland) characterized by medium to high injection viscosity23 was used for augmentation. In the augmented group 2 ml of bone cement were injected at room temperature through each screw without previous filling. Under C-arm visualization cement distribution within the bone was monitored. Full curing of the cement was ensured before testing. For mechanical testing femoral shafts were shortened distally 10 cm below the lesser trochanter and rigidly mounted to a base fixture at 20˚ to the horizontal plane. The shaft was embedded in PMMA (Beracryl, Suter Kunststoffe AG, Fraubrunnen, Switzerland) and fixed to the inclined base.24 A polyethylene cup fixed on an x-y table was used to mimic the acetabulum through which force was applied
Figure 1. Fracture creation at an angle of 45˚ to the horizontal plane to simulate femoral neck fractures as they occur in vivo. A line which was drawn in the middle of the femoral neck, midway between the femoral head cartilage and the intertrochanteric line helped to create the osteotomy.
315
Figure 2. Complete (A) and close-up (B) view of a conventional 7.3 mm cannulated titanium alloy cancellous hip screw perforated with four 2 mm diameter holes, starting 10 mm distal to the tip of the screw over a distance of 10 mm, alternating by 120˚ angles (only two holes are visible on the photographs).
(Fig.3). This setting reflects a 16˚ resultant joint load vector to the vertical plus 4˚ offset of the femoral shaft axis from the sagittal plane.25 Cyclic testing was performed at 2 Hz on a MTS Mini Bionix II 858 hydraulic test system (MTS Systems Corp., Eden Prairie) equipped with a 4 kN load cell. In order to simulate an alternating load during walking, a loading trajectory resulting from in vivo measurements in the human hip was transferred to the femoral head.26
Figure 3. Setup for biomechanical testing. A polyethylene cup fixed on an x-y table was used to mimic the acetabulum through which force was applied. The setting reflects a 16˚ resultant joint load vector to the vertical plus 4˚ offset of the femoral shaft axis from the sagittal plane. JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
316
HOFMANN-FLIRI ET AL.
Starting at 400 N the peak load was monotonically increased at 0.1 N/cycle27 until severe failure of the construct (machine actuator displacement 15 mm). The load valley was maintained at 50 N throughout the test. Failure mode was monitored by taking antero-posterior radiographs every 250 cycles at the load valley (Siemens Arcadic Varic, Siemens Medical Solutions AG, Munich, Germany) and with help of reference pins which were placed in the different bone fragments. Evaluation was done using Matlab (Matlab, Mathworks Inc., Natwick, MA) image processing software. For each test, cycles to failure (15 mm actuator displacement) and cycles to 3 mm vertical head displacement as well as changes in screw angle, cut-out (migration of the screw tips through the femoral head) and telescoping at 8,500 cycles were calculated and statistically evaluated. Statistical analysis was conducted with SPSS 19.0 for Windows (SPSS Inc., Chicago, IL). After checking the data for normal distribution (Shapiro-Wilk Test), paired t-tests were performed to detect significant statistical differences between the two groups regarding BMD, cycles to failure and screw angle, cut-out and telescoping after 8,500 cycles. A survival analysis (Cox-Regression Test) was performed on the cycles to 3 mm head displacement. Retrospective power calculation was performed for statistically significant results. P-values below 0.05 were considered as significant.
RESULTS All values are presented as mean and standard deviation (SD). BMD (in the femoral head) was 212 mgHa/ cm3 (SD 42) for the augmented group and 214 mgHA/ cm3 (SD 43) for the non-augmented group. No statistical difference was found for BMD between the two groups (p ¼ 0.91). Failure (15 mm actuator displacement) occurred in the augmented group at 12,500 cycles (SD 2,480) compared to 15,625 cycles (SD 4,215) in the nonaugmented group (p ¼ 0.041, power 0.593) (Fig. 4). When comparing 3 mm of vertical displacement of the head no significant difference (p ¼ 0.72) was
Figure 4. Graph showing boxplots of the cycles to failure (15 mm actuator displacement) for the augmented and nonaugmented group. JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
detected between the survival curves of the two groups (Fig. 5). At 8,500 load cycles (early onset failure) the augmented group demonstrated a significantly higher change in screw angle of 2.85˚ (SD 0.84) towards varus compared to 1.15˚ (SD 0.93) in the non-augmented group (p ¼ 0.013, power 0.852) (Fig. 6). There was only minimal cut-through (migration of the screw tips inside the femoral head
AO Research Institute Davos, Switzerland, 2Hospital Italiano de Buenos Aires, Argentina, 3Medical University of Innsbruck, Austria
Received 30 April 2015; accepted 24 June 2015 Published online 29 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22978
ABSTRACT: Femoral neck fractures in the elderly are a common problem in orthopedics. Augmentation of screw fixation with bone cement can provide better stability of implants and lower the risk of secondary displacement. This study aimed to investigate whether cement augmentation of three cannulated screws in non-displaced femoral neck fractures could increase implant fixation. A femoral neck fracture was simulated in six paired human cadaveric femora and stabilized with three 7.3 mm cannulated screws. Pairs were divided into two groups: conventional instrumentation versus additional cement augmentation of screw tips with 2 ml TraumacemVþ each. Biomechanical testing was performed by applying cyclic axial load until failure. Failure cycles, axial head displacement, screw angle changes, telescoping and screw cut-out were evaluated. Failure (15 mm actuator displacement) occurred in the augmented group at 12,500 cycles ( 2,480) compared to 15,625 cycles ( 4,215) in the non-augmented group (p ¼ 0.041). When comparing 3 mm vertical displacement of the head no significant difference (p ¼ 0.72) was detected between the survival curves of the two groups. At 8,500 loadcycles (early onset failure) the augmented group demonstrated a change in screw angle of 2.85˚ ( 0.84) compared to 1.15˚ ( 0.93) in the non-augmented group (p ¼ 0.013). The results showed no biomechanical advantage with respect to secondary displacement following augmentation of three cannulated screws in a non-displaced femoral neck fracture. Consequently, the indication for cement augmentation to enhance implant anchorage in osteoporotic bone has to be considered carefully taking into account fracture type, implant selection and biomechanical surrounding. ß 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 34:314–319, 2016. Keywords: femoral neck fracture; implant augmentation; cannulated screw; osteoporosis; PMMA
The annual worldwide incidence of hip fractures exceeds 1.7 million.1 Femoral neck fractures are one of the most frequent, accounting for 60% of all hip fractures and they are becoming more common as the proportion of elderly people increases.2–4 Their association with osteoporotic bone and the risks inherent to the injury and treatment renders their management difficult. One major concern with osteoporotic fracture fixation is related to obtaining secure anchorage of the implant due to the decrease in cortical thickness, cancellous bone density and increased porosity.5,6 The most commonly used implants to fix neck fractures are sliding hip screws and percutaneous cancellous screws. The use of cancellous screws can be a less expensive and less invasive option for non-displaced or minimally displaced fractures, particularly in comparison with total hip arthroplasty.7 However, complications such as secondary displacement are reported in 10–25% and non-union in 7–10% of the cases.3,7,8 Failures after internal fixation usually lead to revision surgery and total hip arthroplasty.3 Bone cement augmentation, as described for different applications in osteoporotic fracture fixation, can
Ladina Hofmann-Fliri and Tomas I. Nicolino contributed equally to this work. Conflict of interest: The authors are not compensated and there are no other institutional subsidies, corporate affiliations, or funding sources supporting this work unless clearly documented and disclosed. Grant sponsor: AOTRAUMA Network; Grant number: AR2008_01. Correspondence to: Markus Windolf (T: þ41-81-414-23-29; F: þ41-81-414-22-88; E-mail: [email protected]) # 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
314
JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
provide better implant stability at reduced risk for secondary displacement. Sermon et al.9 compared polymethylmethacrylate (PMMA) augmented helical blades of a proximal femoral nail with non-augmented blades in human cadaveric femoral heads and found 51% increased load cycles to failure due to application of small amounts of bone cement (3 ml) in the femoral head region. The effect was particularly accentuated in osteoporotic bone. Clinically, it could be shown in a prospective multicenter trial that the same procedure led to good functional results with no complications related to the cement augmentation.10 Other biomechanical studies report similar advantages and, hence, draw a consistent picture of the concept. This includes PMMA augmentation of a sliding hip screw,11,12 of proximal humerus plate screws,13 of distal femur and proximal tibia plate screws,14,15 of lag screws16 and of a tibiotalocalcaneal arthrodesis nail.17 Thus, there is reason to presume that augmenting the tips of cannulated cancellous screws for fixation of femoral neck fractures might improve implant anchorage in a similar manner. The aim of this study was therefore to experimentally evaluate the biomechanical potential of cement augmentation of three cannulated cancellous screws in the femoral head in a non-displaced proximal femoral neck fracture model (AO type 31 B2).
MATERIALS AND METHODS Six pairs of fresh frozen (20˚) human cadaveric femora without soft tissues were provided by the Institute of Pathology from the University Hospital Basel, Switzerland. The mean age was 81 years (minimum 76; maximum 90);
CEMENT AUGMENTATION OF IMPLANTS
three specimen pairs were from female donors and the other three from male donors. Bone mineral density (BMD) was measured in the femoral head for each specimen by peripheral quantitative computed tomography (pQCT) using an XtremeCT (SCANCO Medical AG, Br€ uttisellen, Switzerland). A femoral neck fracture was simulated by an osteotomy in all the specimens. The fracture was created at an angle of 45˚ to the horizontal plane.18 This corresponds to a fracture perpendicular to the axis of the femoral neck if a caputcollum-diaphyseal (CCD) angle of 135˚ is considered.19 A line which was drawn in the middle of the femoral neck, midway between the femoral head cartilage and the intertrochanteric line helped to create the osteotomy (Fig. 1). The fracture was created with an oscillating saw and was stabilized with a triangular configuration of three 7.3 mm cannulated titanium alloy cancellous hip screws (DepuySynthes Inc., Oberdorf, Switzerland) using parallel guide wires. An apex single screw was placed superiorly.20 One from each pair of femora was assigned into one of the two groups: augmented versus conventional screws. In the augmentation group, the cannulated screws were perforated with 4 holes of 2 mm diameter, starting 10 mm distal to the tip of the screw over a distance of 10 mm, alternating by 120˚ angles (Fig. 2). The screws were inserted into the femoral heads up to a distance of 5 mm from the articular surface.21,22 Augmentation was performed through a custom-made side-opening cannula of 2.5 mm diameter. Acrylic bone cement (Traumacem Vþ, DepuySynthes Inc., Oberdorf, Switzerland) characterized by medium to high injection viscosity23 was used for augmentation. In the augmented group 2 ml of bone cement were injected at room temperature through each screw without previous filling. Under C-arm visualization cement distribution within the bone was monitored. Full curing of the cement was ensured before testing. For mechanical testing femoral shafts were shortened distally 10 cm below the lesser trochanter and rigidly mounted to a base fixture at 20˚ to the horizontal plane. The shaft was embedded in PMMA (Beracryl, Suter Kunststoffe AG, Fraubrunnen, Switzerland) and fixed to the inclined base.24 A polyethylene cup fixed on an x-y table was used to mimic the acetabulum through which force was applied
Figure 1. Fracture creation at an angle of 45˚ to the horizontal plane to simulate femoral neck fractures as they occur in vivo. A line which was drawn in the middle of the femoral neck, midway between the femoral head cartilage and the intertrochanteric line helped to create the osteotomy.
315
Figure 2. Complete (A) and close-up (B) view of a conventional 7.3 mm cannulated titanium alloy cancellous hip screw perforated with four 2 mm diameter holes, starting 10 mm distal to the tip of the screw over a distance of 10 mm, alternating by 120˚ angles (only two holes are visible on the photographs).
(Fig.3). This setting reflects a 16˚ resultant joint load vector to the vertical plus 4˚ offset of the femoral shaft axis from the sagittal plane.25 Cyclic testing was performed at 2 Hz on a MTS Mini Bionix II 858 hydraulic test system (MTS Systems Corp., Eden Prairie) equipped with a 4 kN load cell. In order to simulate an alternating load during walking, a loading trajectory resulting from in vivo measurements in the human hip was transferred to the femoral head.26
Figure 3. Setup for biomechanical testing. A polyethylene cup fixed on an x-y table was used to mimic the acetabulum through which force was applied. The setting reflects a 16˚ resultant joint load vector to the vertical plus 4˚ offset of the femoral shaft axis from the sagittal plane. JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
316
HOFMANN-FLIRI ET AL.
Starting at 400 N the peak load was monotonically increased at 0.1 N/cycle27 until severe failure of the construct (machine actuator displacement 15 mm). The load valley was maintained at 50 N throughout the test. Failure mode was monitored by taking antero-posterior radiographs every 250 cycles at the load valley (Siemens Arcadic Varic, Siemens Medical Solutions AG, Munich, Germany) and with help of reference pins which were placed in the different bone fragments. Evaluation was done using Matlab (Matlab, Mathworks Inc., Natwick, MA) image processing software. For each test, cycles to failure (15 mm actuator displacement) and cycles to 3 mm vertical head displacement as well as changes in screw angle, cut-out (migration of the screw tips through the femoral head) and telescoping at 8,500 cycles were calculated and statistically evaluated. Statistical analysis was conducted with SPSS 19.0 for Windows (SPSS Inc., Chicago, IL). After checking the data for normal distribution (Shapiro-Wilk Test), paired t-tests were performed to detect significant statistical differences between the two groups regarding BMD, cycles to failure and screw angle, cut-out and telescoping after 8,500 cycles. A survival analysis (Cox-Regression Test) was performed on the cycles to 3 mm head displacement. Retrospective power calculation was performed for statistically significant results. P-values below 0.05 were considered as significant.
RESULTS All values are presented as mean and standard deviation (SD). BMD (in the femoral head) was 212 mgHa/ cm3 (SD 42) for the augmented group and 214 mgHA/ cm3 (SD 43) for the non-augmented group. No statistical difference was found for BMD between the two groups (p ¼ 0.91). Failure (15 mm actuator displacement) occurred in the augmented group at 12,500 cycles (SD 2,480) compared to 15,625 cycles (SD 4,215) in the nonaugmented group (p ¼ 0.041, power 0.593) (Fig. 4). When comparing 3 mm of vertical displacement of the head no significant difference (p ¼ 0.72) was
Figure 4. Graph showing boxplots of the cycles to failure (15 mm actuator displacement) for the augmented and nonaugmented group. JOURNAL OF ORTHOPAEDIC RESEARCH FEBRUARY 2016
detected between the survival curves of the two groups (Fig. 5). At 8,500 load cycles (early onset failure) the augmented group demonstrated a significantly higher change in screw angle of 2.85˚ (SD 0.84) towards varus compared to 1.15˚ (SD 0.93) in the non-augmented group (p ¼ 0.013, power 0.852) (Fig. 6). There was only minimal cut-through (migration of the screw tips inside the femoral head
