1. Biomechanical study
1.1 Implants
The experimental work will be undertaken in the Laboratory of Technology and Strength of Materials at the Department of Mechanical Engineering and Aeronautics of Patras University. Two implant systems will be compared against biomechanical testing, the classic Gamma3 and the InterTan nail (IT).
The Gamma3®nail (Stryker, Mahwah, NJ, USA) that would be used in the current study has a length of 180 mm, 11 mm nail diameter and 130° neck-shaft angle and utilizes a single lag screw with a standard diameter of 10.5 mm. The Trigen Intertan (Smith+Nephew, Watford, UK) nail is a current generation IM device, featuring a dual lag screw configuration comprised of a 11.0 mm superior lag screw and a 7.0 mm smaller integrated screw that allows for linear compression of the fragments at the fracture site while providing high rotational stability [27,28]. The proximal part of the nail has a trapezoid design that provides a pressfit fixation in the metaphyseal region thus positioning more material on the lateral side of the nail where tensile/stretching forces tend to be greatest. Finally, the “clothes pin” distal tip provides less rigidity to decrease the stress riser and reduce the incidence of anterior thigh pain. An 11 mm nail with a 130ο neck-shaft angle will be used for the present study.
We would hypothesize that the above variations, may create differences in strain distribution and thus dissimilar biomechanical behaviors of these two nails. The main hypothesis of our biomechanical study is that the InterTan nail would provide higher stiffness and load to failure and less varus collapse in unstable intertrochanteric fractures (AO/OTA A2). The transmission of forces as measured by cortical surface-strain distribution in the proximal femur will be evaluated using digital image correlation (DIC) and classic strain gauges.
1.2 Preparation of specimens
Eight synthetic composite osteoporotic femurs of medium size (model: 3503; SawBones Worldwide, WA, USA) with 10 PCF low-density cancellous, thin-walled low-density cortical shell and 16 mm hollow canal, would be randomly assigned to two groups (n=4 in each group) to receive either a classic Gamma3 or an IT nail. Two additional specimens will be kept intact and will be served as reference for system calibration and material properties validation. The use of sawbone models with validated mechanical properties [37-39] was preferred instead of cadaveric bones to eliminate size, geometry and mechanical variability between specimens thus increasing stanchness of our numerical analysis due to the availability of the finite element 3D digital file with coded material properties of our specimen (Femur, Finite Element Model of #3503, SawBones Worldwide, WA, USA).
The instrumentation of Gamma3 and IT will be applied in a standard manner, according to the manufacturer’s manual, first on the intact sawbones (prior to fracture creation) under image intensifier using antero- posterior and lateral imaging to ensure proper implant position and a tip-to-apex distance (TAD) < 20 mm (Fig 1a-c).
The instrumentation will be removed and the fractures would be created using an electric saw and a custom-made cutting guide. For the creation of an unstable intertrochanteric fracture (AO/OTA 31A2.2) the fracture line will be at an angle of 47° to the horizontal level running from the lesser trochanter followed by removal of an additional wedge of bone, containing the lesser trochanter. The sawbones will be re-instrumented and prepared for biomechanical testing (Fig 1d,e).
The natural inclination of the femur at the single-leg stance, which is 11o of abduction in the frontal plane and neutral on the sagittal plane will be ensured using the surface of the posterior condyle as reference point [40] (Fig 2a). The femurs will then be fixed distally at the supracondylar level to a custom resin mold (Smooth-Cast 300 Series) properly reinforced with an orthogonal steel plate (Fig 2b). Standard strain gauges (model C2A-06-125LW-350, 350𝛺±0.6%, –50° up to +80°C) will be applied at the medial and lateral side of the specimen 2 cm below the fracture line as well as 2 cm below the distal interlocking screw (Fig 2c). Finally, a thin layer of white paint with black speckles (Fig 2d) will be applied at the proximal end of the femur to measure the transmission of forces, in the cortical surface-strain distribution using digital image correlation (DIC). (Fig 2e).
1.3 Loading scheme
Once the implants were settled well and distal ends were embedded, the two groups of model bones were fixed on the biomechanical test machine (MTS 647, hydraulic wedge grip) (Fig 2f). Initially, all specimens will be subjected to an axial pressure of 100-200 N for 10 cycles with a loading rate of 1 Hz to eliminate creeping effect.
Axial cycling load compression will be applied to simulate the stress experienced by a patient with 70 kg body weight at 4–6 weeks postoperatively [41, 42]. Firstly, the axial compression test is carried out under cyclic loading, with an initial load of 400 N, incremental load of 100 N, and maximum load set to 1400 N, divided into 10 subgroups; each subgroup will be conducted in a cycle of 10000 times, respectively, with a loading rate of 1 Hz. After the cyclic test, the average fracture gap movements will be recorded. A torsional test will follow with the following parameters: Starting from 0°, the maximum torsion angle will be set to 3° with a loading rate of 0.1°/s. Torque at the angle of 1°, 2°, 3° will be recorded, respectively. Finally, an axial compression failure test will be performed with a loading rate of 4.6 mm/s continuously, until fatigue failure; the latter is defined as fracture gap > 20 mm, nail cutting‐out or breakage and fracture line found near the distal locking screw.
1.4 Statistical analysis
The SPSS 19.0 statistical software package will produce the statistical analysis of the study. With the aid of Shapiro-Wilk test, the normal deviation of cycles to failure, failure load and axial construct stiffness in each group will be monitored. To screen the fluctuation in the parameters of interest, the Levene test will be utilized. Any major difference between the two implants will be registered with paired-samples t test, while we set a P=.05 as level of significance.
2. Finite Element Analysis (FEA)
Finite element analysis of the implanted sawbones will be performed using the ANSYS v.19.2 commercial FE code (ANSYS inc. Cannonsburg, Pennsylvania, USA). The femoral geometry model will be obtained by the manufacturer (3D CAD digital file with coded material properties, Sawbones, Vashon Island, Washington, USA). The geometrical features of each implant (Gamma3 and IT) will be characterized following 3D CT scanning. Only the basic geometrical features will be modeled using FEA, since it has been recognized that several additional details (i.e., threads of the lag screw(s) or distal interlocking screw) significantly complicate the analysis, without affecting considerably the computed stress and strain fields [43].
Sawbones geometry will be imported in the ANSYS FE code and the fracture plane will be modelled by creating a proper surface that divide the cortical and cancellous volumes of the bone for the unstable fracture cases. The native neck shaft angle and the anatomic reduction angle will be 135°. A central position of the hip screw with a TAD of 16–18 mm and a minimum of 2 mm width of cancellous bone between the hip screw and the femoral cortex will be applied. Several FEA models will be created with the aim to study different scenarios of reduction (in varus or valgus) as well as inappropriate lag screw positions as illustrated in Figure 3. Our hypothesis is that even with inappropriate reduction or lag screw(s) mispositioning the biomechanical properties of IT will be better than Gamma3.
As has been proposed in a similar study [43], the developed geometries will be meshed using tetrahedral elements with an optimal element size of 3 mm as this leads to acceptable results regarding accuracy, without significantly increasing the computational cost. Appropriate linear elastic, isotropic and homogeneous material properties will be assigned to the several parts of the FE model, based on the data provided by the manufacturers. The longitudinal Young’s modulus for the cortical bone (82pcf fiber filled epoxy) is Ecortical = 5.2 GPa, and for the cancellous bone (10 pcf solid foam) Ecancellous = 58 MPa, while for the Gamma3 and IT the titanium alloy the Young’s modulus is Ematerial = 116 GPa. For the modeling of the distal femoral fixation during the experiments, fully fixed boundary conditions will be applied on the FE models surface nodes situated at the lower 130 mm of the femur model. A constant vertical force of 1000 N will be applied on the top of the femoral head; as a linear-elastic FE analysis will be considered, the calculated results will be derived for other load values using linear interpolation or extrapolation. Linear regression modeling will be utilized in order to associate a set of independent variables or predictors (implant type, lag screw(s) positioning and reduction angle) with any measurement outcome, also referred to as the dependent variable or target, separately. For each model, the beta estimates and the p-values will be returned, among others. Statistical significance was taken when p < 0.05.
3. Clinical study
Design
This study will be a prospective, randomized clinical trial to compare the clinical and radiological outcome between single screw chephalomedullary nailing (Gamma3) and integrated dual interlocking nailing (InterTan -IT) for the treatment of unstable (31A21-3) intertrochanteric fractures. The paper complies with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 Statement for reporting of clinical trial protocols. The study is sponsored by the University of Patras Research Committee (ELKE) (University of Patras Campus, Rio, Greece). Ethical approval has been already acquired by the Ethics Committee of the University Hospital of Patras (approval number: 3373/7-2-2022) and the patients’ written consent will be obtained before participating in the study. The study is listed with the International Standard Randomized Controlled Trial Number (ISRCTN): https://doi.org/10.1186/ISRCTN15588442 (registered on 19/4/2022).
Primary objective
The primary aim of this study is to investigate if there are any significant differences in the main radiological measurements (cut-out, varus displacement, loss of reduction, malunion, nonunion) between Gamma3 and IT in closed unstable intertrochanteric fractures (31A21-3) in patients aged over 70 years-old at the 24-week follow-up evaluation.
The second primary aim would be the functional status at the 24-week follow up using patient reported disease specific scores (Harris Hip score HHS-and Oxford hip score-OHS).
Secondary objectives (Table 1)
Several secondary objectives will also be studied to evaluate the effectiveness and safety of the devices by quantifying and drawing presumptions from observed differences between treatment groups in the following:
- Comparison of perioperative and intraoperative surgical data (age-adjusted Carlson Comorbidity Index [44], operation time, fluoroscopy time and dosage, blood loss, length of hospital stay, prescription of pain-killers, osteoporosis assessment using postoperative DEXA of the unaffected hip, union time and intraoperative surgeon related complications, including lag screw malposition, propagation of the fracture, non-anatomical reduction, varus/valgus deformity, rotational deformity and tip-apex distance-TAD).
- Pain level at the perioperative period and at 6-, 12-, and 24-weeks postoperatively using the visual analogue scale (VAS) Pain Score.
- Patient reported general health status prior to surgery and at 6-, 12- and 24-weeks post-surgery using the SF-36 form, the EQ-5D-3L Questionnaire, the SARC-F Index and the Elderly Mobility Scale (EMS).
Patient selection
The patients that will take part in the study, are those over 70 years old, admitted in the Orthopaedic Clinic of the University Hospital of Patras, Greece, suffering from a closed unstable (31A21-3) intertrochanteric fracture. Those patients will be randomized in two groups (Gamma3 and IT). The pre-operative data will be comprised of demographic information, radiological examination of the hips and other questionnaires to conclude the overall physical health and mobility prior the fracture. The participants will be free to withdraw from the study or the investigator can withdraw a patient. If a participant withdraws, they will be listed to a “Change of Status” form and will be invited to present for an endpoint follow-up.
Eligibility criteria
- Closed intertrochanteric fracture (AO/OTA:31A2)
- Age over 70 years
- Presentation to hospital within 7 days of injury
- No concomitant injuries or prior operations to the unaffected hip.
Exclusion criteria
- Concomitant injuries affecting treatment or rehabilitation of the affected limb
- Associated neurovascular injuries requiring immediate surgery.
- Patients with limited Greek proficiency including family members
- Patients where consent is refused
- Patients with severe dementia, non-ambulated and with severe associated diseases prohibiting operative intervention.
Sample size
This study will use two primary clinical outcome measures namely the Oxford Hip Score and the Harris Hip Score. The minimum clinically important differences (MCIDs) for the HHS have been estimated between 7 and 10, whereas for the OHS between 5 and 7. The aim will be to recruit 78 patients in each group as this will provide sufficient participants to obtain a power of 90% for both primary clinical outcome measures. With an allowance for 15% drop-out, the total number of patients required will be 194 (97 in each group). If recruitment proves to be problematic during the course of the trial, then the target will be lowered and the more usual 80% power level will be considered sufficient. For this scenario, the total number of patients required will be 140 (including 15% for drop-out).
Randomization
Randomization sequence will be created using Stata 9.0 (StataCorp, College Station, TX) statistical software and will be stratified with a 1:1 allocation ratio using random block sizes of 2, 4, and 6. An investigator not involved in the main study will perform the randomization process. These data will be stored on computer and will be available at the day of the intervention.
Concealment mechanism
For each surgical session patient details and the randomized treatment will be sealed in an opaque envelope that will be given to the coordinated nurse just before the start of the intervention. The patients, relatives, investigators, nurses, and all relevant personnel would be blinded to the indented treatment. The envelope will be opened by the coordinated nurse after the patient is sedated and placed to the radiological table, the hip is properly prepared and the surgeons are ready to apply the intervention. Both instrumentations for Gamma3 or IT would be available at the operative room. After opening the envelope, the procedure will be unblinded for all participants.
Surgical intervention
The patients for the study will be randomly divided in two groups. Group A: Gamma3 nail (Stryker) and Group B: InterTan (IT) nail (Smith & Nephew). The surgery will take place in one specific operating room with the standardized surgical technique, using the same c-arm for the intra-operative radiological evaluation. Anatomical reduction of the fracture will be intended in all cases prior to nail insertion. If this is not possible, several techniques of closed or open reduction will be applied in order to achieve an acceptable fixation of the fracture. The patients’ position will be on the traction table, using a minimal approach. The lag screw(s) in both groups will be put in the center of the femoral head in the lateral and anteroposterior X-ray view, while the Tip to Apex Distance (TAD) will be approximately 20 mm. All participants in both groups will receive the same physical therapy and adequate anti-osteoporotic treatment.
Radiographic evaluation
A standard technique will be kept, in order to have the same parameters in all radiographs. For the antero-posterior ones of the pelvis, both legs of the patients will be rotated inwards 15o. The radiographic measurements will be monitored by two distinguished examiners through RadiAnt DICOM Viewer. Intra-class correlation coefficients (ICC) will check the inter- and intra-rater reliability. The parameters that will be radiographically evaluated are the TAD, the position of the chephalomedulary nail in the femur head and the alignment [45]. Moreover, throughout the follow-up the varus malalignment of the implant will be assessed, as well as the cut-out of the nail and the percentage of nonunion or malunion. Varus collapse is a rotation of the proximal part of the femur anteriorly for at least 3o.
Outcome assessment
- Patient reported general health status prior to surgery and at 6-, 12- and 24-weeks post-surgery using the SF-36 form, the EQ-5D-3L Questionnaire, the SARC-F Index and the Elderly Mobility Scale (EMS).
- Pain level at the perioperative period and at 6, 12 and 24 weeks postoperatively using the visual analogue scale (VAS) Pain Score.
- Comparison of perioperative and intraoperative surgical data (delay for surgery, age-adjusted Carlson Comorbidity Index, operation time, fluoroscopy time and dosage, blood loss, length of hospital stay, prescription of pain-killers, osteoporosis assessment using postoperative DEXA of the unaffected hip, union time and intraoperative surgeon related complications, including lag screw malposition, propagation of the fracture, non-anatomical reduction, varus/valgus deformity, rotational deformity and tip-apex distance (TAD) using postoperative CT of the pelvis).
- Patients’ functional status at 24 weeks follow up using the patient reported disease specific scores (Harris Hip score HHS and Oxford hip score-OHS).
Table 1. Timeline chart of patient clinical evaluation
Outcome measure
|
Specification
|
Score
|
Assessment times
Pre-op Post-op 6w 12w 24w
|
Patient-reported outcome measures
|
|
|
|
|
|
EQ-5D-3L
|
General health status
|
5-15
|
√
|
|
√
|
√
|
√
|
SF-36
|
General health status
|
0-100
|
√
|
|
√
|
√
|
√
|
Elderly Mobility Scale
|
Fragility Assessment
|
0-20
|
√
|
|
√
|
√
|
√
|
SARC-F Questionnaire
|
Sarcopenia assessment
|
0-10
|
√
|
|
√
|
√
|
√
|
|
|
|
|
|
|
|
|
NPRS
|
Joint pain
|
0-10
|
√
|
√
|
√
|
√
|
√
|
|
|
|
|
|
|
|
|
Hip Joint-specific measures
|
Harris Hip Score (HHS)
|
Joint specific score
|
0-100
|
√
|
|
√
|
√
|
√
|
Oxford Hip Score (OHS)
|
Joint specific score
|
0-100
|
√
|
|
√
|
√
|
√
|
|
Surgeon and patient related complications
|
|
|
Intra-op
|
Post-op
|
√
|
√
|
√
|
|
|
|
|
|
|
|
|
Failure rates
|
|
|
|
Post-op
|
√
|
√
|
√
|
NRPS: Numerical Pain Rating Scale, SF-36: Short-Form 36, EMS: Elderly Mobility Scale, SARC-F: Strength, assistance with walking, rising from a chair, climbing stairs, and falls questionnaire, HHS: Harris Hip Score, OHS: Oxford Hip Score
Follow-up assessments
All the patients should undergo a clinical and radiological follow-up at 6-, 12- and 24 weeks postoperatively. The follow-up will be concluded by antero-posterior radiographs of the pelvis, lateral radiography of the treated hip and questionnaire like HHS, OHS, SF-36, EQ-5D-3L, EMS, Sarc-f, NPRS scale and Vas score.
Statistical analysis
The primary outcome measure of device radiological failure is considered a binary outcome (device failed/did not fail). A binary logistic regression model will be performed to assess the association between the outcome of device failure and the predictor of device type (Gamma3, Intertan). The complication rate of G3 vs InterTan at 24 weeks post-operatively would be compared using a chi-squared (at the 5% level). The differences between HHS and OHS between groups will be assessed using an independent samples t-test at 24 weeks postoperatively at the 5% level. Test levels, will be adjusted using the methods of Holm-Bonferroni to allow for the multiple comparisons. A linear regression analysis will also be used to quantify the effects of the treatment groups on each of the primary outcome measures, after adjusting for the effects of a range of other important, potentially confounding, factors (e.g. age, gender) recorded for each patient.