- Email: [email protected]
In vivo comparison of hip mechanics for minimally invasive versus traditional total hip arthroplasty
Available online at www.sciencedirect.com
Clinical Biomechanics 23 (2008) 127–134 www.elsevier.com/locate/clinbiomech
Clinical Biomechanics Award 2007
In vivo comparison of hip mechanics for minimally invasive versus traditional total hip arthroplasty Diana Glaser a
a,*
, Douglas A. Dennis
a,b
, Richard D. Komistek a, Todd M. Miner
b
University of Tennessee, Biomedical Engineering Department, 301 Perkins Hall, Knoxville, TN 37996, USA b Colorado Joint Replacement, 2535 South Downing Street Suite 100, Denver, CO 80210, USA Received 30 July 2007; accepted 24 September 2007
Abstract Background. Minimally invasive surgery has been developed to reduce incision length, muscle damage, and rehabilitation time. However, reduced exposure of anatomical landmarks may result in technical errors and inferior implant survivorship. The objective of this study was to compare in vivo motions and hip joint contact forces during gait in total hip arthroplasty subjects, performed with either minimally invasive surgery or standard surgical approaches. Methods. Fifteen subjects implanted using either minimally invasive surgery anterolateral, minimally invasive surgery posterolateral, or traditional posterolateral total hip arthroplasty were evaluated using fluoroscopy while performing gait on a treadmill. Kinematics, obtained using 3D-to-2D image registration technique, were input as temporal functions in a 3D inverse dynamic mathematical model that determines in vivo soft tissue and hip contact forces. Findings. The subjects implanted with posterolateral and anterolateral minimally invasive surgery demonstrated significantly less separation than those implanted with the traditional approach (P < 0.01). The minimally invasive surgery subjects also experienced lower average maximum peak forces, with 3.2 body weight for the anterolateral minimally invasive surgery and 2.9 body weight for the posterolateral minimally invasive surgery subjects, compared to 3.5 body weight for the traditional subjects (P = 0.02 and P = 0.03, respectively). Interpretation. This is the first study to compare in vivo weight-bearing kinematics, separation and kinetics for traditional, anterolateral minimally invasive surgery and posterolateral minimally invasive surgery total hip arthroplasty subject groups. Our data indicated in all analyzed parameters differences between the minimally invasive surgery and the traditional groups, with favorable results for the minimally invasive surgery subjects. This may be related, to a reduction in stabilizing soft tissues after a minimally invasive surgery procedure, leading to lower bearing surface forces at the femoral head–acetabular cup interface. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: MIS; THA; Kinematics; Kinetics; Separation; In vivo; Anterolateral THA; Posterolateral THA; Forces; Fluoroscopy; Gait; Hip; Mathematical modeling; Kane’s dynamics
1. Introduction Total hip arthroplasty (THA) is now a predictable and highly effective procedure routinely performed throughout the world. Improvements in outcomes of THA have been attributed to advances in implant design, fixation methods, prophylactic therapies, preadmission education, and reha*
Corresponding author. E-mail address: [email protected] (D. Glaser).
0268-0033/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2007.09.015
bilitation methods (Mardones et al., 2005). Adequate exposure and good visualization of anatomical landmarks are crucial for the success of the THA, which traditionally have been performed utilizing incisions of up to 40 cm in length (Bottner et al., 2006). Minimally invasive surgery (MIS) has become popular because of its potential to reduce soft tissue damage and because of the complimentary benefits. The most often reported advantages of MIS include accelerated rehabilitation, improved cosmetic appearance, less pain, shorter hospitalization, lower risk of complications,
128
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
decreased surgical time, and decreased blood loss (Chung et al., 2004; Wenz et al., 2002; Pavone et al., 2001; Pagnano et al., 2007a,b). MIS THA is viewed by its supporters as a logical evolution of less invasive methods. Conversely, concerns with issues relating to reduced visualization, such as implant malposition, neurovascular injury, poor implant fixation, or compromised long-term results, have become a debate (Berry et al., 2003). Current reports on MIS THA have primarily focused on early functional results, complication rates, amount of blood loss, rehabilitation time, implant position and severity of pain (Bottner et al., 2006; Chung et al., 2004; Howell et al., 2004; Pagnano et al., 2007a). In most cases, the comparison and performance evaluation are based on questionnaires, clinical examination, and radiographic review (Swanson and Hanna, 2003). Previous gait analyses of subjects implanted with THA have focused on external hip joint angles and the time-distance parameters of speed, stride length, single leg stance time and cadence (Fiedler et al., 2006; Pagnano et al., 2007b; Sirianni et al., 2007). Fluoroscopic studies confirmed that the femoral head slides away from the acetabular component, often referred to as hip separation, during gait and when performing a hip abduction/adduction activity (Dennis et al., 2001; Komistek et al., 2002; Northcut, 1998). No known research has been performed to evaluate the kinematics and kinetics of subjects with traditional and MIS THAs during in vivo, weight-bearing activities. However, such a comparison is useful in identifying the variance and success of the subjects’ performance after THA. For example, altered hip loading during weight-bearing activities may result in increased contact and muscle forces, leading to increased separation combined with pain and accelerated wear. Thus, it is logical to assume that gait mechanics would demonstrate a stronger correlation and serve as a better method in the evaluation of THA performance, when compared to other variables previously studied. However, the literature lacks well-designed studies that provide objective evidence of the superiority of MIS compared to traditional THA in major parameters such as kinematics and kinetics. For that reason, the purpose of this study was to examine the role of hip kinematics and kinetics on walking performance in subjects after THA and to compare the effectiveness of MIS approaches to a traditional THA technique. 2. Methods 2.1. Subjects Fifteen subjects were analyzed under in vivo, weightbearing conditions using video fluoroscopy. All subjects received a hybrid metal-on-polyethylene hip replacement with a femoral head diameter of either 28 or 32 mm, performed using a single incision and by a single, experienced, fellowship-trained surgeon. Three groups were analyzed based on the surgical approach: five patients underwent THA using a traditional approach (>15 cm, Aufranc and
Head, 1977), five patients using anterolateral MIS (ALMIS; Berger, 2004) and five patients using posterolateral MIS (PL-MIS; Bottner et al., 2006; Chung et al., 2004). Subjects in each group were matched for age, height, weight, body mass index (BMI), diagnosis and femoral head diameter to control for confounding variables possibly having influence on the hip performance and gait kinematics. The patient demographics are listed in detail in Table 1. Approval was obtained from the Institutional Review Board at The University of Tennessee and subjects signed an informed consent form prior to participation. All THA patients with excellent clinical results without pain, functional deficits, or generalized inflammation, were invited to participate in the study (Harris hip scores HHS > 90 points; Harris and Sledge, 1990). All subjects were afflicted with only unilateral hip arthritis, and demonstrated no evidence of any postoperative hip subluxation or dislocation. No patient walked with a detectable limp and all could actively abduct their operated hip against gravity without difficulty. These inclusion criteria were necessary to assure mobility and to eliminate errors related to the fact that patient rehabilitation was incomplete. Such factors could affect gait patterns and limit the conclusions that can be drawn from the study. An attempt was made to restore femoral offset and orient the acetabular component within the safe zone position (Lewinnek et al., 1978) for all THA patients. The average postoperative follow-up durations at the time of analysis were 6.4 months (3–12 months), 4.7 months (3.5–4.5 months) and 3.7 months (2.3–6.5 months) for patients implanted using a traditional, AL-MIS and PL-MIS approach, respectively; further, these values were not significantly different. The length of the skin incision, operating time, total estimated intra-operative blood loss, postoperative narcotic requirements (as a measure of pain following the procedure), and length of hospitalization were also recorded for each group. 2.2. Surgical procedures The traditional THA utilized an incision greater than 15 cm. The fascia lata (gluteus maximus fascia) was split over the length of the skin incision. The posterior capsule and short external rotators, including the piriformis, superior and inferior gemilli, obturator internus, and a portion of the proximal aspect of the quadratus femoris muscles, were released and anatomically repaired after implantation. No special retractors or acetabular reamers were exploited. In subjects in whom a MIS approach was utilized, the incision length was reduced by about 40% compared to the traditional THA. The PL-MIS was performed with an averaged skin length of roughly 9 cm. The fascia lata was split over the same length. The posterior capsule and short external rotator muscles were incised from just superior to the piriformis to the superior aspect of the quadratus femoris muscle. The gluteus maximus tendon and quadratus femoris were left intact. The hip exposure was accomplished using the small skin incision as a mobile win-
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
129
Table 1 Patient demographics Subject #
Procedure
Sex
Leg side
Diagnosis
Age (years)
Post-OP time (months)
Height (in)
Weight (lb)
BMI
Hospital stay (days)
THA THA THA THA THA
M M F M F
R R L R R
OA OA OA OA OA
56 76 75 67 59
6 4 12 4 12
67 68.5 64 63 63
208 175 167 128 158
32.6 26.2 28.7 22.7 28.0
4 3 3 4 4
6 7 8 9 10
AL-MIS AL-MIS AL-MIS AL-MIS AL-MIS
F F M M F
R R R R L
OA OA AVN OA OA
66 42 54 64 69
6.5 6 4.5 2 4
60 67 68 73 60
128 136 163 190 126
25.0 21.3 24.8 25.1 24.6
3 2 2 2 3
11 12 13 14 15
PL-MIS PL-MIS PL-MIS PL-MIS PL-MIS
F M M M M
R R L L L
OA OA OA OA OA
77 65 72 64 67
4.5 3.5 3.5 3.5 3.5
65 66 72 71 73
113 140 195 190 215
18.8 22.6 26.4 26.5 28.4
3 3 2 2 3
1 2 3 4 5
THA: traditional THA. PL-MIS: posterolateral minimally invasive THA. AL-MIS: anterolateral minimally invasive THA. OA: osteoarthritis. AVN: avascular necrosis. Leg side: R = right, L = left. Sex: M = male, F = female.
dow (Chung et al., 2004; Bottner et al., 2006). Special lighted retractors were utilized for enhanced visibility, and acetabular reamers with ‘side cut-outs’ were employed to allow easier passage of the reamers into the smaller wound. After implantation, the short external rotators and posterior capsule were repaired anatomically with heavy, non-absorbable suture. In the AL-MIS, the skin and fascia lata incision were similarly performed as described for the PL-MIS. Then, the anterior 30% of the gluteus medius and minimus were released, and a limited anterior subtotal capsulectomy was performed. The gluteus medius and minimus tendons as well as the remaining anterior capsule were repaired anatomically after implantation. 2.3. Data collection A preliminary screening questionnaire was used for evaluation of the HHS. In the present study, subjects were analyzed during gait, as it is the predominant weight-bearing activity and is, therefore, the most important aspect of evaluating patient function. Each subject performed normal walking while on a level treadmill. A two-camera system and fluoroscopic unit were used to capture the in vivo weight-bearing movement of the hip joint implants as well as the leg motion during the gait activity. A single stride beginning from toe-off to the subsequent toe-off was analyzed for each patient. Data were examined using a threedimensional (3D) (Dennis et al., 1996) registration process and MATLABTM (The MathWorks, Inc., MA, USA) applications written at the Center for Musculoskeletal Research, Tennessee (Mahfouz et al., 2003). Individual fluoroscopic video frames at specified time intervals were
digitized. The measured variables included the femoral and pelvic 3D rotational and translational kinematics. The relative motions of the femur and pelvis were calculated based on the transformation matrices of each body and were subsequently used to determine the distance between the center of the femoral head and the acetabular component (Fig. 1). This was then applied to diagnose if femoral head sliding (separation) of the femoral head from the acetabular component had occurred (Dennis et al., 2001). 2.4. Error analysis A previously published error analysis verified the accuracy of the 3D model-fitting process (Dennis et al., 1996; Sarojak, 1998). An error threshold of 0.5 mm was obtained; therefore, femoral head sliding was predicted to occur if the femoral head-to-acetabular cup distance was greater than 0.5 mm.
Fig. 1. Approach for measuring the separation between femoral head and acetabular component. D = femoral head movement (separation).
130
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
2.5. Mathematical modeling The mathematical model of the lower extremity was defined as a set of differential equations that represents the dynamics of the system. Newton’s law was used to define the model; however, Kane’s dynamics (Kane and Levinson, 1985, 1996) was applied to simplify the mathematical model and to optimize its performance. In contrast to the use of the classical formulation of dynamics, Kane’s unique method of auxiliary generalized speeds builds compact, customized and computationally efficient ordinary differential equations. For application to larger multi-body systems, the symbolic manipulator AutolevTM (OnLine Dynamics, Inc., Sunnyvale, CA, USA; Kane and Levinson, 1996) was used. The developed mathematical model is based on an inverse dynamics model of the lower extremities (Fig. 2), which provides the three-dimensional interaction resultant forces and moments. Those kinetic values represent the vector sums of forces in muscles, ligaments, and capsules, forces on the joint contact surface, and the moments engendered by those forces. There are more unknowns than equations of motion, which makes the problem indeterminate; therefore, the model was simplified by grouping functionally similar hip muscles, applying the reduction technique. Details on the method of calculating the hip contact forces have been reported previously (Komistek et al., 1994, 1998; Dennis et al., 1996). Additions were made by including the key components of the quadriceps mechanism: patella ligament, quadriceps muscles and the patello-femoral interactive forces. Also integrated in the model were the hip capsular ligaments, which were modeled as springs. The iliofemoral ligament was built out of two fibers and the ischiofemoral ligament out of six fibers. The forces of the hip capsular ligaments were calculated based on their change of length, which was obtained from the relative position of the bodies using position vectors. A
study done by Hewitt et al. (2002) quantified the values of the stiffness coefficients for the hip capsular ligaments via tensile tests on cadaveric specimens, which were subsequently utilized in the present model. A left human leg from our extensive computer tomography scan data collection, which was similar to the size and shape of the subject tested, was used to determine bone mass centers, as well as the origin and insertion position of the ligaments and muscles. Segment inertia parameters were obtained from a study done by De Leva (1996). The relative joint kinematics obtained from fluoroscopy and the ground reaction force data were curve-fit using piecewise functions to an R > 0.99 and were entered into the mathematical model as input functions. The developed model gives an accurate 3D representation of the in vivo kinematics and kinetics for subjects after THA for gait, ensuring that the soft tissues and interactive forces were modeled correctly. The calculated hip contact forces were compared to the in vivo measured forces produced by Bergman’s telemetric hip data (Bergmann, 2001). A comparison between the traditional as a control group, PLMIS, and AL-MIS patients was performed to determine which group leads to the most desirable conditions, postoperatively. 2.6. Statistical analysis The principal analysis of this study was a comparison of the separation and force values for patients implanted with a traditional, Al-MIS, or PL-MIS surgical approach. Student’s t-test with a risk level (a-level) of 0.05, which is especially adequate in estimating the mean of a population when the sample size is small, was used to test for significance between the groups. The selected samples were chosen to be independent of each other (e.g., individuals were randomly selected for each group). The statistical analyses were conducted using JMP (V6, SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Kinematics
Fig. 2. Schematic diagram including bodies, coordinate systems and contact points.
From fluoroscopy, the obtained relative transformations represent the sequential rotation about the three orthogonal axes and displacement transfers starting from an initial position. The sequence order of the rotations is 3, 2, 1 with the axes defined in Fig. 2. Rotation around the 3-axis represents pure flexion/extension. Rotation around the 2-axis is here denoted internal/external rotation, although it must be noted that this rotation is around the transformed 2axis. Similarly, the rotation around the 1-axis is not a pure abduction/adduction, but rather a rotation around the twice rotated 1-axis. The average rotations and their variance between the subjects in each group are displayed in Figs. 3–5 for the traditional, PL-MIS and AL-MIS patients, respectively. All groups experienced similar mag-
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Fig. 3. Average (solid lines) and variance (shaded areas) relative rotations for traditional patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
131
Five of five subjects (100%) having a THA implanted using a traditional THA experienced separation within the acetabular component greater than our threshold error of 0.5 mm. Only one of five subjects (20%) implanted using the AL-MIS and two of five subjects (40%) implanted using the PL-MIS approach experienced greater than 0.5 mm of separation. The separation values for each subject are summarized in Table 2. The average separation distance detected in subjects following a traditional approach was 0.83 mm (0.66–0.9 mm). The average values for subjects implanted using AL-MIS and PL-MIS approaches were 0.5 mm (0.35–0.68 mm) and 0.48 mm (0.38–0.64 mm), respectively. There was a significant difference between the traditional and the PL-MIS groups (P = 0.002) as well as between the traditional and the AL-MIS groups (P < 0.001). No significant difference was found between the two MIS approach groups. 3.2. Hip bearing contacts forces
Fig. 4. Average (solid lines) and variance (shaded areas) relative rotations for PL-MIS patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
The hip bearing contact forces for the patients in the traditional and both MIS groups are displayed in Fig. 6, whereas variance in all groups is displayed in Fig. 7. The maximum peak forces as well as the forces at different stages of the gait cycle for each subject are summarized in Table 2. The maximum peak force of 4.1 BW was achieved by a traditional study group patient, while the AL-MIS and PLMIS groups attained only 3.6 and 3.3 BW, respectively. The average maximum peak force for the THA patients implanted using a traditional surgical approach was 3.5 BW, while that for the MIS groups was 3.2 and 2.9 BW for the AL (P = 0.03) and PL approaches (P = 0.02), respectively. The PL-MIS group showed a significant lower forces than the traditional group (P = 0.04) at heel-strike. The average force for the terminal stance was significant lower for the PL-MIS versus traditional group (P = 0.03). For all the stages of the gait cycle, the MIS subjects experienced lower hip forces. A differentiation between both MIS approaches showed that the PL-MIS approach led to slightly lower forces than the AL-MIS approach, although this was not statistically significant. 4. Discussion
Fig. 5. Average (solid lines) and variance (shaded areas) relative rotations for AL-MIS patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
nitudes and pattern for all rotations. However, the variance for the traditional group was significant greater than that for either MIS group (P < 0.05).
The hip joint kinematics and kinetics of patients who underwent traditional, AL-MIS and PL-MIS THAs were investigated during early postoperative gait, and the influence of the surgical approach on those parameters was evaluated. The investigated approaches likely affect the structures around the hip (soft tissues such as capsular ligaments and myotendinous units) in unequal magnitude, which led to the conclusive presumption that those groups would have distinct postoperative gait characteristics. Since the traditional approach requires an increased magnitude of surgical dissection, it was hypothesized that this approach would have a greater affect on the gait kinematics and result in higher hip separation values and hip contact forces, when
132
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Table 2 Kinematic and kinetic results Subject #
Procedure
Separation
Forces (BW) HS
LS
TS
MIN (stance)
MAX
THA THA THA THA THA
0.66 0.89 0.83 0.90 0.89
1.63 1.57 1.63 0.86 1.92
0.58 0.58 1.39 0.25 0.05
1.52 2.71 1.41 0.72 1.50
3.79 3.06 3.11 2.57 3.24
3.71 4.08 2.30 2.90 3.55
1.43 1.26 1.48 0.70 2.09
3.79 4.11 3.11 2.91 3.55
6 7 8 9 10
AL-MIS AL-MIS AL-MIS AL-MIS AL-MIS
0.49 0.48 0.39 0.64 0.44
1.22 2.00 1.15 2.49 1.92
0.01 0.29 0.48 0.08 0.04
1.18 1.49 1.25 0.61 1.55
3.01 3.62 2.52 2.73 2.11
2.75 3.43 2.13 3.15 3.54
1.9 2.4 0.9 2.2 1.6
3.01 3.62 2.52 3.15 3.54
11 12 13 14 15
PL-MIS PL-MIS PL-MIS PL-MIS PL-MIS
0.36 0.68 0.46 0.35 0.65
0.54 1.59 1.59 1.26 1.54
0.14 0.34 0.31 -0.52 0.21
1.03 1.16 0.87 0.74 0.56
2.64 2.87 3.00 3.28 2.67
2.73 2.34 2.56 2.71 2.34
0.56 1.57 1.21 1.16 1.56
2.73 2.87 3.00 3.28 2.67
TO (t = 0) 1 2 3 4 5
SW (min)
BW: body weight. TO: toe-off. SW: swing phase. MS: mid stance. HS: heel-strike. LS: loading response (approx. 33% of stance phase). TS: terminal stance (approx. 66% of stance phase).
Fig. 6. Average hip forces (BW) for traditional (solid line), PL-MIS (solid line with markers) and AL-MIS (markers only) patients.
Fig. 7. Variance hip forces for traditional, posterolateral and anterolateral MIS Patients.
compared to a MIS approach. This hypothesis was confirmed by the results of the present investigation. MIS has been hotly discussed in recent years, but conclusions about the safety and efficacy of these methods are not yet clear. Some previous studies have found reduced blood loss, acceptable morbidity, satisfactory radiographic results, and a faster recovery time for subjects implanted with a procedure involving use of a smaller incision (Sculco, 2003; Berger, 2003; Bottner et al., 2006). This agrees with our findings, as both MIS groups experienced a shorter duration of hospitalization (Table 1). Another comparative study of THA using traditional versus MIS reported significant improvements in limp, distance walked, and ability to climb stairs in the MIS group at
six months following the operative procedure (DiGioia, 2003). Bottner et al. (2006) suggests that MIS approaches are reasonable approaches to utilize in the performance of THA. Femoral head sliding, often referred as separation, was analyzed with the expectation to reflect differences between the approaches in the opening of the hip joint capsule, as well as in the amount of muscle damage. The hip joint is often described as a ball-socket joint; however, the true motion of the joint deviates when femoral head sliding occurs. During separation, the large, uniform contact area between femoral head and acetabulum shifts to undesirable edge loading conditions between both components. A significantly higher incidence and magnitude of hip separation
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
was observed in the traditional surgical approach group. Only limited studies have been performed on calculating in vivo hip joint contact forces including additional muscle and ligaments forces (Callaghan et al., 1992; Brand et al., 1994). Previous studies have been performed using computer models to predict muscle and joint reaction forces, but these methods have not been successfully validated against in vivo data (Winter, 1990). Therefore, the goal of the current study was to develop a mathematical model for the hip joint that accurately represents the in vivo loading conditions of THA patients and, at the same time, consider the relationship between muscle and joint forces. A comparison of the current results to those in a telemetric hip reported by Bergmann (2001) demonstrated that the force patterns and the magnitudes were similar. The average predicted forces in our analysis were 3.5, 3.2, and 2.9 BW for the traditional, AL-MIS, and PL-MIS subjects, respectively, while Bergmann reported forces from 2.3 to 3.5 BW under similar conditions. In the present study, subjects implanted with a MIS technique experienced significantly lower hip force magnitudes than the traditional subjects. Additionally, the variance of the kinematics and kinetics among the individuals within the groups was higher for the traditional subjects group and lowest for the patients implanted with PL-MIS. This may indicate that MIS procedures lead to more consistent and reproducible results. In summary, our data indicates in all analyzed parameters differences between the MIS groups and those implanted with THA using a traditional surgical approach, with more favorable results obtained in the MIS subjects. No differences were observed between the two MIS procedures. The reduced incidence and magnitude of femoral head sliding as well as the lower hip force magnitudes observed in MIS subjects is likely related, at least in part, to a reduction in disruption of supporting soft tissue structures when compared to disruption magnitudes of performing these procedures using a traditional approach. The limitations of the presented study are the short-term follow-up time and the small number of patients in each of the groups that were studied. It will be important to conduct follow-up studies in the future that involve a greater number of subjects within each group, more surgeons, differing surgical techniques and longer postoperative periods. Also, only one activity was analyzed; therefore, it cannot be definitively concluded that the results from this study in gait will reflect other activities. We realize that in addition to the surgical approach, supplementary factors such as BMI, implant positioning and rehabilitation techniques may play a role in hip mechanics. In the present study, an attempt was made to control for multiple variables such as BMI, age, femoral head diameter, diagnosis, success of the surgery (HHS, no subluxation, dislocation or limp), as well as component orientation. Those variables were matched between the groups, so that effects of those unwanted factors on soft tissue support were minimized to provide a clear investigation of the desired analyzed parameters. This fact, along with the single, experienced
133
surgeon involved in the study, build up the strengths of our study. Fluoroscopy has proven to be a precise method for obtaining in vivo, weight-bearing kinematics and determination of the presence of hip joint separation. The mathematical model created for this study has shown results which are in agreement with those of telemetric studies. Thus, the accurate procedures involved in this work enable extensive evaluation of gait characteristics as they have not been studied previously. In summary, this is the first study to compare in vivo, weight-bearing kinematics and kinetics for individuals following either a traditional, AL-MIS or PL-MIS surgical approach. Our study demonstrates evidence that the use of MIS techniques in performing THA may provide short-term advantages as well as implant longevity if the superior mechanics are maintained over longer intervals. 5. Conflict of Interest Financial support for the study has been obtained from Zimmer, Inc., Warsaw, IN, USA. Except for the funding, the sponsors did not play a role in study design, data collection, analysis and data interpretation. Acknowledgments The authors are thankful to Zimmer, Inc., Warsaw, IN, USA for providing financial support. The authors would also like to acknowledge the contributions of Phat D. Nguyen and Stuart Deaderick. References Aufranc, O.T., Head, W.C., 1977. A versatile surgical approach to the hip. Clin. Orthop. Relat. Res. 128 (October), 285–286. Berger, R.A., 2003. The technique and early results of minimally invasive total hip arthroplasty. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Berger, R.A., 2004. Mini-incision total hip replacement using an anterolateral approach: technique and results. Orthop. Clin. North. Am. 35 (2), 143–151. Bergmann, G., 2001. Hip98, loading of the hip joint: contact forces, gait patterns, muscle forces, activities. J. Biomech. 34 (7) (cd-rom attachment). Berry, D.J., Berger, R.A., Callaghan, J.J., et al., 2003. Minimally invasive total hip arthroplasty. Development, early results and a critical analysis. J. Bone Joint Surg. 85A, 2235–2246. Bottner, F., Delgado, S., Sculco, T.P., 2006. Minimally invasive total hip replacement: the posterolateral approach. Am. J. Orthop. 35 (5), 218– 224. Brand, R.A., Pedersen, D.R., Davy, D.T., et al., 1994. Comparison of hip force calculations and measurements in the same patient. J. Arthroplasty 9, 45–51. Callaghan, J.J., Fulghum, C.S., Glisson, R.R., Stranne, S.K., 1992. The effect of femoral stem geometry on interface motion in uncemented porous-coated total hip prosthesis. J. Bone Joint Surg. 74, 839–848. Chung, W.K., Liu, D., Foo, L.S.S., 2004. Mini-incision total hip replacement-surgical technique and early results. J. Orthop. Surg. 12, 19–24. De Leva, P., 1996. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J. Biomech. 29, 1223–1230.
134
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Dennis, D.A., Komistek, R.D., Hoff, W.A., Gabriel, S., 1996. In vivo knee kinematics derived using an inverse perspective technique. Clin. Orthop. Relat. Res. 331, 107–117. Dennis, D.A., Komistek, R.D., Northcut, E.J., et al., 2001. In vivo determination of hip joint separation and the forces generated due to impact loading conditions. J. Biomech. 34, 623–629. DiGioia III, A.M., 2003. The role of navigational tools in more accurate and less invasive THA. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA, Symposium F. Fiedler, E., Hildebrand, M., Moisio, K.C., Wimmer, M.A., et al., 2006. MIS approach to THA affects gait differences. In: American Academy of Orthopaedic Surgeons AAOS Meeting, Chicago, IL. Harris, W.H., Sledge, C.B., 1990. Total hip and total knee replacement. N. Engl. J. Med., 323–725. Hewitt, J.D., Glisson, R.R., Guilak, A.K., et al., 2002. The mechanical properties of the human hip capsule ligaments. J. Arthroplasty 17, 82– 89. Howell, J.R., Masri, B.A., Duncan, C.P., 2004. Minimally invasive versus standard incision anterolateral hip replacement: a comparative study. Orthop. Clin. North. Am. 35, 153–162. Kane, T.R., Levinson, D.A., 1985. Dynamics: Theory and Applications. McGraw-Hill, NY. Kane, T.R., Levinson, D.A., 1996. Dynamics Online: Theory and Implementation with Autolev, Online Dynamics, Sunnyvale, CA. Komistek, R.D., Stiehl, J.B., Paxson, R.D., 1994. Mathematical model of the human lower extremity. In: Vossoughi, J. (Ed.), Biomedical Engineering Recent Developments, Washington DC, pp. 1098–1101. Komistek, R.D., Stiel, J.B., Dennis, D.A., et al., 1998. Mathematical model of the lower extremity joint reaction forces using Kane’s method of dynamics. J. Biomech. 31, 185–189. Komistek, R.D., Dennis, D.A., Ochoa, J.A., et al., 2002. In vivo comparison of hip separation after metal-on-metal or metal-on polyethylene total hip arthroplasty. J. Bone Joint Surg. 84A, 1836–1841. Lewinnek, G.E., Lewis, J.L., Tarr, R., et al., 1978. Incidence of dislocation following hip arthroplasty for patients in the rehabilitation setting. J. Bone Joint Surg. Am. 60, 217–220.
Mahfouz, M.R., Hoff, W.A., Komistek, R.D., Dennis, D.A., 2003. A robust method for registration of three-dimensional knee implant models to two-dimensional fluoroscopy images. IEEE Trans. Med. Imaging 22, 1561–1574. Mardones, R., Pagnano, M., Nemanich, J., et al., 2005. Muscle damage after total hip arthroplasty done with the two-incision and mini-posterior techniques. Clin. Orthop. Relat. Res. 441, 63– 67. Northcut, E.J., 1998. In vivo determination of hip joint separation. Colorado school of mines. Engineering Thesis. Pagnano, M.W., Leone, J., Meneghini, R.M., et al., 2007a. A prospective randomized trial shows two-incision THAs do not recover quicker than mini-posterior THAs. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Pagnano, M.W., Meneghini, R.M., Kaufman, K.R., et al., 2007b. No benefit of the two-incision THA over mini-posterior THA: a comprehensive gait analysis study. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Pavone, V., Chimemto, G., Sharrock, N., Sculco, T., 2001. The role of incision length in total hip arthroplasty. J. Bone Joint Surg. 83B (Suppl 2), 213S. Sarojak, M.E., 1998. Model-fit: an interactive pose determining system. Colorado school of mines. Engineering Thesis, Golden, CO. Sculco, T.P., 2003. Advances in minimally invasive surgery. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Sirianni, L.A., Dorr, L.D., Yun, A., et al., 2007. Gait analysis of MIS THR. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Swanson, T.V., Hanna, R.S., 2003. Advantages of cementless THA using minimally invasive surgical technique. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Wenz, J.F., Gurkan, I., Jibodh, S.R., 2002. Mini-incision total hip arthroplasty: a comparative assessment of perioperative outcomes. Orthopedics 25, 1031–1043. Winter, D.A., 1990. Biomechanics and Motor Control of Human Movement. John Wiley & Sons, Inc., Canada.
Clinical Biomechanics 23 (2008) 127–134 www.elsevier.com/locate/clinbiomech
Clinical Biomechanics Award 2007
In vivo comparison of hip mechanics for minimally invasive versus traditional total hip arthroplasty Diana Glaser a
a,*
, Douglas A. Dennis
a,b
, Richard D. Komistek a, Todd M. Miner
b
University of Tennessee, Biomedical Engineering Department, 301 Perkins Hall, Knoxville, TN 37996, USA b Colorado Joint Replacement, 2535 South Downing Street Suite 100, Denver, CO 80210, USA Received 30 July 2007; accepted 24 September 2007
Abstract Background. Minimally invasive surgery has been developed to reduce incision length, muscle damage, and rehabilitation time. However, reduced exposure of anatomical landmarks may result in technical errors and inferior implant survivorship. The objective of this study was to compare in vivo motions and hip joint contact forces during gait in total hip arthroplasty subjects, performed with either minimally invasive surgery or standard surgical approaches. Methods. Fifteen subjects implanted using either minimally invasive surgery anterolateral, minimally invasive surgery posterolateral, or traditional posterolateral total hip arthroplasty were evaluated using fluoroscopy while performing gait on a treadmill. Kinematics, obtained using 3D-to-2D image registration technique, were input as temporal functions in a 3D inverse dynamic mathematical model that determines in vivo soft tissue and hip contact forces. Findings. The subjects implanted with posterolateral and anterolateral minimally invasive surgery demonstrated significantly less separation than those implanted with the traditional approach (P < 0.01). The minimally invasive surgery subjects also experienced lower average maximum peak forces, with 3.2 body weight for the anterolateral minimally invasive surgery and 2.9 body weight for the posterolateral minimally invasive surgery subjects, compared to 3.5 body weight for the traditional subjects (P = 0.02 and P = 0.03, respectively). Interpretation. This is the first study to compare in vivo weight-bearing kinematics, separation and kinetics for traditional, anterolateral minimally invasive surgery and posterolateral minimally invasive surgery total hip arthroplasty subject groups. Our data indicated in all analyzed parameters differences between the minimally invasive surgery and the traditional groups, with favorable results for the minimally invasive surgery subjects. This may be related, to a reduction in stabilizing soft tissues after a minimally invasive surgery procedure, leading to lower bearing surface forces at the femoral head–acetabular cup interface. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: MIS; THA; Kinematics; Kinetics; Separation; In vivo; Anterolateral THA; Posterolateral THA; Forces; Fluoroscopy; Gait; Hip; Mathematical modeling; Kane’s dynamics
1. Introduction Total hip arthroplasty (THA) is now a predictable and highly effective procedure routinely performed throughout the world. Improvements in outcomes of THA have been attributed to advances in implant design, fixation methods, prophylactic therapies, preadmission education, and reha*
Corresponding author. E-mail address: [email protected] (D. Glaser).
0268-0033/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2007.09.015
bilitation methods (Mardones et al., 2005). Adequate exposure and good visualization of anatomical landmarks are crucial for the success of the THA, which traditionally have been performed utilizing incisions of up to 40 cm in length (Bottner et al., 2006). Minimally invasive surgery (MIS) has become popular because of its potential to reduce soft tissue damage and because of the complimentary benefits. The most often reported advantages of MIS include accelerated rehabilitation, improved cosmetic appearance, less pain, shorter hospitalization, lower risk of complications,
128
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
decreased surgical time, and decreased blood loss (Chung et al., 2004; Wenz et al., 2002; Pavone et al., 2001; Pagnano et al., 2007a,b). MIS THA is viewed by its supporters as a logical evolution of less invasive methods. Conversely, concerns with issues relating to reduced visualization, such as implant malposition, neurovascular injury, poor implant fixation, or compromised long-term results, have become a debate (Berry et al., 2003). Current reports on MIS THA have primarily focused on early functional results, complication rates, amount of blood loss, rehabilitation time, implant position and severity of pain (Bottner et al., 2006; Chung et al., 2004; Howell et al., 2004; Pagnano et al., 2007a). In most cases, the comparison and performance evaluation are based on questionnaires, clinical examination, and radiographic review (Swanson and Hanna, 2003). Previous gait analyses of subjects implanted with THA have focused on external hip joint angles and the time-distance parameters of speed, stride length, single leg stance time and cadence (Fiedler et al., 2006; Pagnano et al., 2007b; Sirianni et al., 2007). Fluoroscopic studies confirmed that the femoral head slides away from the acetabular component, often referred to as hip separation, during gait and when performing a hip abduction/adduction activity (Dennis et al., 2001; Komistek et al., 2002; Northcut, 1998). No known research has been performed to evaluate the kinematics and kinetics of subjects with traditional and MIS THAs during in vivo, weight-bearing activities. However, such a comparison is useful in identifying the variance and success of the subjects’ performance after THA. For example, altered hip loading during weight-bearing activities may result in increased contact and muscle forces, leading to increased separation combined with pain and accelerated wear. Thus, it is logical to assume that gait mechanics would demonstrate a stronger correlation and serve as a better method in the evaluation of THA performance, when compared to other variables previously studied. However, the literature lacks well-designed studies that provide objective evidence of the superiority of MIS compared to traditional THA in major parameters such as kinematics and kinetics. For that reason, the purpose of this study was to examine the role of hip kinematics and kinetics on walking performance in subjects after THA and to compare the effectiveness of MIS approaches to a traditional THA technique. 2. Methods 2.1. Subjects Fifteen subjects were analyzed under in vivo, weightbearing conditions using video fluoroscopy. All subjects received a hybrid metal-on-polyethylene hip replacement with a femoral head diameter of either 28 or 32 mm, performed using a single incision and by a single, experienced, fellowship-trained surgeon. Three groups were analyzed based on the surgical approach: five patients underwent THA using a traditional approach (>15 cm, Aufranc and
Head, 1977), five patients using anterolateral MIS (ALMIS; Berger, 2004) and five patients using posterolateral MIS (PL-MIS; Bottner et al., 2006; Chung et al., 2004). Subjects in each group were matched for age, height, weight, body mass index (BMI), diagnosis and femoral head diameter to control for confounding variables possibly having influence on the hip performance and gait kinematics. The patient demographics are listed in detail in Table 1. Approval was obtained from the Institutional Review Board at The University of Tennessee and subjects signed an informed consent form prior to participation. All THA patients with excellent clinical results without pain, functional deficits, or generalized inflammation, were invited to participate in the study (Harris hip scores HHS > 90 points; Harris and Sledge, 1990). All subjects were afflicted with only unilateral hip arthritis, and demonstrated no evidence of any postoperative hip subluxation or dislocation. No patient walked with a detectable limp and all could actively abduct their operated hip against gravity without difficulty. These inclusion criteria were necessary to assure mobility and to eliminate errors related to the fact that patient rehabilitation was incomplete. Such factors could affect gait patterns and limit the conclusions that can be drawn from the study. An attempt was made to restore femoral offset and orient the acetabular component within the safe zone position (Lewinnek et al., 1978) for all THA patients. The average postoperative follow-up durations at the time of analysis were 6.4 months (3–12 months), 4.7 months (3.5–4.5 months) and 3.7 months (2.3–6.5 months) for patients implanted using a traditional, AL-MIS and PL-MIS approach, respectively; further, these values were not significantly different. The length of the skin incision, operating time, total estimated intra-operative blood loss, postoperative narcotic requirements (as a measure of pain following the procedure), and length of hospitalization were also recorded for each group. 2.2. Surgical procedures The traditional THA utilized an incision greater than 15 cm. The fascia lata (gluteus maximus fascia) was split over the length of the skin incision. The posterior capsule and short external rotators, including the piriformis, superior and inferior gemilli, obturator internus, and a portion of the proximal aspect of the quadratus femoris muscles, were released and anatomically repaired after implantation. No special retractors or acetabular reamers were exploited. In subjects in whom a MIS approach was utilized, the incision length was reduced by about 40% compared to the traditional THA. The PL-MIS was performed with an averaged skin length of roughly 9 cm. The fascia lata was split over the same length. The posterior capsule and short external rotator muscles were incised from just superior to the piriformis to the superior aspect of the quadratus femoris muscle. The gluteus maximus tendon and quadratus femoris were left intact. The hip exposure was accomplished using the small skin incision as a mobile win-
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
129
Table 1 Patient demographics Subject #
Procedure
Sex
Leg side
Diagnosis
Age (years)
Post-OP time (months)
Height (in)
Weight (lb)
BMI
Hospital stay (days)
THA THA THA THA THA
M M F M F
R R L R R
OA OA OA OA OA
56 76 75 67 59
6 4 12 4 12
67 68.5 64 63 63
208 175 167 128 158
32.6 26.2 28.7 22.7 28.0
4 3 3 4 4
6 7 8 9 10
AL-MIS AL-MIS AL-MIS AL-MIS AL-MIS
F F M M F
R R R R L
OA OA AVN OA OA
66 42 54 64 69
6.5 6 4.5 2 4
60 67 68 73 60
128 136 163 190 126
25.0 21.3 24.8 25.1 24.6
3 2 2 2 3
11 12 13 14 15
PL-MIS PL-MIS PL-MIS PL-MIS PL-MIS
F M M M M
R R L L L
OA OA OA OA OA
77 65 72 64 67
4.5 3.5 3.5 3.5 3.5
65 66 72 71 73
113 140 195 190 215
18.8 22.6 26.4 26.5 28.4
3 3 2 2 3
1 2 3 4 5
THA: traditional THA. PL-MIS: posterolateral minimally invasive THA. AL-MIS: anterolateral minimally invasive THA. OA: osteoarthritis. AVN: avascular necrosis. Leg side: R = right, L = left. Sex: M = male, F = female.
dow (Chung et al., 2004; Bottner et al., 2006). Special lighted retractors were utilized for enhanced visibility, and acetabular reamers with ‘side cut-outs’ were employed to allow easier passage of the reamers into the smaller wound. After implantation, the short external rotators and posterior capsule were repaired anatomically with heavy, non-absorbable suture. In the AL-MIS, the skin and fascia lata incision were similarly performed as described for the PL-MIS. Then, the anterior 30% of the gluteus medius and minimus were released, and a limited anterior subtotal capsulectomy was performed. The gluteus medius and minimus tendons as well as the remaining anterior capsule were repaired anatomically after implantation. 2.3. Data collection A preliminary screening questionnaire was used for evaluation of the HHS. In the present study, subjects were analyzed during gait, as it is the predominant weight-bearing activity and is, therefore, the most important aspect of evaluating patient function. Each subject performed normal walking while on a level treadmill. A two-camera system and fluoroscopic unit were used to capture the in vivo weight-bearing movement of the hip joint implants as well as the leg motion during the gait activity. A single stride beginning from toe-off to the subsequent toe-off was analyzed for each patient. Data were examined using a threedimensional (3D) (Dennis et al., 1996) registration process and MATLABTM (The MathWorks, Inc., MA, USA) applications written at the Center for Musculoskeletal Research, Tennessee (Mahfouz et al., 2003). Individual fluoroscopic video frames at specified time intervals were
digitized. The measured variables included the femoral and pelvic 3D rotational and translational kinematics. The relative motions of the femur and pelvis were calculated based on the transformation matrices of each body and were subsequently used to determine the distance between the center of the femoral head and the acetabular component (Fig. 1). This was then applied to diagnose if femoral head sliding (separation) of the femoral head from the acetabular component had occurred (Dennis et al., 2001). 2.4. Error analysis A previously published error analysis verified the accuracy of the 3D model-fitting process (Dennis et al., 1996; Sarojak, 1998). An error threshold of 0.5 mm was obtained; therefore, femoral head sliding was predicted to occur if the femoral head-to-acetabular cup distance was greater than 0.5 mm.
Fig. 1. Approach for measuring the separation between femoral head and acetabular component. D = femoral head movement (separation).
130
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
2.5. Mathematical modeling The mathematical model of the lower extremity was defined as a set of differential equations that represents the dynamics of the system. Newton’s law was used to define the model; however, Kane’s dynamics (Kane and Levinson, 1985, 1996) was applied to simplify the mathematical model and to optimize its performance. In contrast to the use of the classical formulation of dynamics, Kane’s unique method of auxiliary generalized speeds builds compact, customized and computationally efficient ordinary differential equations. For application to larger multi-body systems, the symbolic manipulator AutolevTM (OnLine Dynamics, Inc., Sunnyvale, CA, USA; Kane and Levinson, 1996) was used. The developed mathematical model is based on an inverse dynamics model of the lower extremities (Fig. 2), which provides the three-dimensional interaction resultant forces and moments. Those kinetic values represent the vector sums of forces in muscles, ligaments, and capsules, forces on the joint contact surface, and the moments engendered by those forces. There are more unknowns than equations of motion, which makes the problem indeterminate; therefore, the model was simplified by grouping functionally similar hip muscles, applying the reduction technique. Details on the method of calculating the hip contact forces have been reported previously (Komistek et al., 1994, 1998; Dennis et al., 1996). Additions were made by including the key components of the quadriceps mechanism: patella ligament, quadriceps muscles and the patello-femoral interactive forces. Also integrated in the model were the hip capsular ligaments, which were modeled as springs. The iliofemoral ligament was built out of two fibers and the ischiofemoral ligament out of six fibers. The forces of the hip capsular ligaments were calculated based on their change of length, which was obtained from the relative position of the bodies using position vectors. A
study done by Hewitt et al. (2002) quantified the values of the stiffness coefficients for the hip capsular ligaments via tensile tests on cadaveric specimens, which were subsequently utilized in the present model. A left human leg from our extensive computer tomography scan data collection, which was similar to the size and shape of the subject tested, was used to determine bone mass centers, as well as the origin and insertion position of the ligaments and muscles. Segment inertia parameters were obtained from a study done by De Leva (1996). The relative joint kinematics obtained from fluoroscopy and the ground reaction force data were curve-fit using piecewise functions to an R > 0.99 and were entered into the mathematical model as input functions. The developed model gives an accurate 3D representation of the in vivo kinematics and kinetics for subjects after THA for gait, ensuring that the soft tissues and interactive forces were modeled correctly. The calculated hip contact forces were compared to the in vivo measured forces produced by Bergman’s telemetric hip data (Bergmann, 2001). A comparison between the traditional as a control group, PLMIS, and AL-MIS patients was performed to determine which group leads to the most desirable conditions, postoperatively. 2.6. Statistical analysis The principal analysis of this study was a comparison of the separation and force values for patients implanted with a traditional, Al-MIS, or PL-MIS surgical approach. Student’s t-test with a risk level (a-level) of 0.05, which is especially adequate in estimating the mean of a population when the sample size is small, was used to test for significance between the groups. The selected samples were chosen to be independent of each other (e.g., individuals were randomly selected for each group). The statistical analyses were conducted using JMP (V6, SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Kinematics
Fig. 2. Schematic diagram including bodies, coordinate systems and contact points.
From fluoroscopy, the obtained relative transformations represent the sequential rotation about the three orthogonal axes and displacement transfers starting from an initial position. The sequence order of the rotations is 3, 2, 1 with the axes defined in Fig. 2. Rotation around the 3-axis represents pure flexion/extension. Rotation around the 2-axis is here denoted internal/external rotation, although it must be noted that this rotation is around the transformed 2axis. Similarly, the rotation around the 1-axis is not a pure abduction/adduction, but rather a rotation around the twice rotated 1-axis. The average rotations and their variance between the subjects in each group are displayed in Figs. 3–5 for the traditional, PL-MIS and AL-MIS patients, respectively. All groups experienced similar mag-
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Fig. 3. Average (solid lines) and variance (shaded areas) relative rotations for traditional patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
131
Five of five subjects (100%) having a THA implanted using a traditional THA experienced separation within the acetabular component greater than our threshold error of 0.5 mm. Only one of five subjects (20%) implanted using the AL-MIS and two of five subjects (40%) implanted using the PL-MIS approach experienced greater than 0.5 mm of separation. The separation values for each subject are summarized in Table 2. The average separation distance detected in subjects following a traditional approach was 0.83 mm (0.66–0.9 mm). The average values for subjects implanted using AL-MIS and PL-MIS approaches were 0.5 mm (0.35–0.68 mm) and 0.48 mm (0.38–0.64 mm), respectively. There was a significant difference between the traditional and the PL-MIS groups (P = 0.002) as well as between the traditional and the AL-MIS groups (P < 0.001). No significant difference was found between the two MIS approach groups. 3.2. Hip bearing contacts forces
Fig. 4. Average (solid lines) and variance (shaded areas) relative rotations for PL-MIS patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
The hip bearing contact forces for the patients in the traditional and both MIS groups are displayed in Fig. 6, whereas variance in all groups is displayed in Fig. 7. The maximum peak forces as well as the forces at different stages of the gait cycle for each subject are summarized in Table 2. The maximum peak force of 4.1 BW was achieved by a traditional study group patient, while the AL-MIS and PLMIS groups attained only 3.6 and 3.3 BW, respectively. The average maximum peak force for the THA patients implanted using a traditional surgical approach was 3.5 BW, while that for the MIS groups was 3.2 and 2.9 BW for the AL (P = 0.03) and PL approaches (P = 0.02), respectively. The PL-MIS group showed a significant lower forces than the traditional group (P = 0.04) at heel-strike. The average force for the terminal stance was significant lower for the PL-MIS versus traditional group (P = 0.03). For all the stages of the gait cycle, the MIS subjects experienced lower hip forces. A differentiation between both MIS approaches showed that the PL-MIS approach led to slightly lower forces than the AL-MIS approach, although this was not statistically significant. 4. Discussion
Fig. 5. Average (solid lines) and variance (shaded areas) relative rotations for AL-MIS patients (flexion, external rotation, adduction are positive values; extension, internal rotations, abduction are negative values).
nitudes and pattern for all rotations. However, the variance for the traditional group was significant greater than that for either MIS group (P < 0.05).
The hip joint kinematics and kinetics of patients who underwent traditional, AL-MIS and PL-MIS THAs were investigated during early postoperative gait, and the influence of the surgical approach on those parameters was evaluated. The investigated approaches likely affect the structures around the hip (soft tissues such as capsular ligaments and myotendinous units) in unequal magnitude, which led to the conclusive presumption that those groups would have distinct postoperative gait characteristics. Since the traditional approach requires an increased magnitude of surgical dissection, it was hypothesized that this approach would have a greater affect on the gait kinematics and result in higher hip separation values and hip contact forces, when
132
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Table 2 Kinematic and kinetic results Subject #
Procedure
Separation
Forces (BW) HS
LS
TS
MIN (stance)
MAX
THA THA THA THA THA
0.66 0.89 0.83 0.90 0.89
1.63 1.57 1.63 0.86 1.92
0.58 0.58 1.39 0.25 0.05
1.52 2.71 1.41 0.72 1.50
3.79 3.06 3.11 2.57 3.24
3.71 4.08 2.30 2.90 3.55
1.43 1.26 1.48 0.70 2.09
3.79 4.11 3.11 2.91 3.55
6 7 8 9 10
AL-MIS AL-MIS AL-MIS AL-MIS AL-MIS
0.49 0.48 0.39 0.64 0.44
1.22 2.00 1.15 2.49 1.92
0.01 0.29 0.48 0.08 0.04
1.18 1.49 1.25 0.61 1.55
3.01 3.62 2.52 2.73 2.11
2.75 3.43 2.13 3.15 3.54
1.9 2.4 0.9 2.2 1.6
3.01 3.62 2.52 3.15 3.54
11 12 13 14 15
PL-MIS PL-MIS PL-MIS PL-MIS PL-MIS
0.36 0.68 0.46 0.35 0.65
0.54 1.59 1.59 1.26 1.54
0.14 0.34 0.31 -0.52 0.21
1.03 1.16 0.87 0.74 0.56
2.64 2.87 3.00 3.28 2.67
2.73 2.34 2.56 2.71 2.34
0.56 1.57 1.21 1.16 1.56
2.73 2.87 3.00 3.28 2.67
TO (t = 0) 1 2 3 4 5
SW (min)
BW: body weight. TO: toe-off. SW: swing phase. MS: mid stance. HS: heel-strike. LS: loading response (approx. 33% of stance phase). TS: terminal stance (approx. 66% of stance phase).
Fig. 6. Average hip forces (BW) for traditional (solid line), PL-MIS (solid line with markers) and AL-MIS (markers only) patients.
Fig. 7. Variance hip forces for traditional, posterolateral and anterolateral MIS Patients.
compared to a MIS approach. This hypothesis was confirmed by the results of the present investigation. MIS has been hotly discussed in recent years, but conclusions about the safety and efficacy of these methods are not yet clear. Some previous studies have found reduced blood loss, acceptable morbidity, satisfactory radiographic results, and a faster recovery time for subjects implanted with a procedure involving use of a smaller incision (Sculco, 2003; Berger, 2003; Bottner et al., 2006). This agrees with our findings, as both MIS groups experienced a shorter duration of hospitalization (Table 1). Another comparative study of THA using traditional versus MIS reported significant improvements in limp, distance walked, and ability to climb stairs in the MIS group at
six months following the operative procedure (DiGioia, 2003). Bottner et al. (2006) suggests that MIS approaches are reasonable approaches to utilize in the performance of THA. Femoral head sliding, often referred as separation, was analyzed with the expectation to reflect differences between the approaches in the opening of the hip joint capsule, as well as in the amount of muscle damage. The hip joint is often described as a ball-socket joint; however, the true motion of the joint deviates when femoral head sliding occurs. During separation, the large, uniform contact area between femoral head and acetabulum shifts to undesirable edge loading conditions between both components. A significantly higher incidence and magnitude of hip separation
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
was observed in the traditional surgical approach group. Only limited studies have been performed on calculating in vivo hip joint contact forces including additional muscle and ligaments forces (Callaghan et al., 1992; Brand et al., 1994). Previous studies have been performed using computer models to predict muscle and joint reaction forces, but these methods have not been successfully validated against in vivo data (Winter, 1990). Therefore, the goal of the current study was to develop a mathematical model for the hip joint that accurately represents the in vivo loading conditions of THA patients and, at the same time, consider the relationship between muscle and joint forces. A comparison of the current results to those in a telemetric hip reported by Bergmann (2001) demonstrated that the force patterns and the magnitudes were similar. The average predicted forces in our analysis were 3.5, 3.2, and 2.9 BW for the traditional, AL-MIS, and PL-MIS subjects, respectively, while Bergmann reported forces from 2.3 to 3.5 BW under similar conditions. In the present study, subjects implanted with a MIS technique experienced significantly lower hip force magnitudes than the traditional subjects. Additionally, the variance of the kinematics and kinetics among the individuals within the groups was higher for the traditional subjects group and lowest for the patients implanted with PL-MIS. This may indicate that MIS procedures lead to more consistent and reproducible results. In summary, our data indicates in all analyzed parameters differences between the MIS groups and those implanted with THA using a traditional surgical approach, with more favorable results obtained in the MIS subjects. No differences were observed between the two MIS procedures. The reduced incidence and magnitude of femoral head sliding as well as the lower hip force magnitudes observed in MIS subjects is likely related, at least in part, to a reduction in disruption of supporting soft tissue structures when compared to disruption magnitudes of performing these procedures using a traditional approach. The limitations of the presented study are the short-term follow-up time and the small number of patients in each of the groups that were studied. It will be important to conduct follow-up studies in the future that involve a greater number of subjects within each group, more surgeons, differing surgical techniques and longer postoperative periods. Also, only one activity was analyzed; therefore, it cannot be definitively concluded that the results from this study in gait will reflect other activities. We realize that in addition to the surgical approach, supplementary factors such as BMI, implant positioning and rehabilitation techniques may play a role in hip mechanics. In the present study, an attempt was made to control for multiple variables such as BMI, age, femoral head diameter, diagnosis, success of the surgery (HHS, no subluxation, dislocation or limp), as well as component orientation. Those variables were matched between the groups, so that effects of those unwanted factors on soft tissue support were minimized to provide a clear investigation of the desired analyzed parameters. This fact, along with the single, experienced
133
surgeon involved in the study, build up the strengths of our study. Fluoroscopy has proven to be a precise method for obtaining in vivo, weight-bearing kinematics and determination of the presence of hip joint separation. The mathematical model created for this study has shown results which are in agreement with those of telemetric studies. Thus, the accurate procedures involved in this work enable extensive evaluation of gait characteristics as they have not been studied previously. In summary, this is the first study to compare in vivo, weight-bearing kinematics and kinetics for individuals following either a traditional, AL-MIS or PL-MIS surgical approach. Our study demonstrates evidence that the use of MIS techniques in performing THA may provide short-term advantages as well as implant longevity if the superior mechanics are maintained over longer intervals. 5. Conflict of Interest Financial support for the study has been obtained from Zimmer, Inc., Warsaw, IN, USA. Except for the funding, the sponsors did not play a role in study design, data collection, analysis and data interpretation. Acknowledgments The authors are thankful to Zimmer, Inc., Warsaw, IN, USA for providing financial support. The authors would also like to acknowledge the contributions of Phat D. Nguyen and Stuart Deaderick. References Aufranc, O.T., Head, W.C., 1977. A versatile surgical approach to the hip. Clin. Orthop. Relat. Res. 128 (October), 285–286. Berger, R.A., 2003. The technique and early results of minimally invasive total hip arthroplasty. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Berger, R.A., 2004. Mini-incision total hip replacement using an anterolateral approach: technique and results. Orthop. Clin. North. Am. 35 (2), 143–151. Bergmann, G., 2001. Hip98, loading of the hip joint: contact forces, gait patterns, muscle forces, activities. J. Biomech. 34 (7) (cd-rom attachment). Berry, D.J., Berger, R.A., Callaghan, J.J., et al., 2003. Minimally invasive total hip arthroplasty. Development, early results and a critical analysis. J. Bone Joint Surg. 85A, 2235–2246. Bottner, F., Delgado, S., Sculco, T.P., 2006. Minimally invasive total hip replacement: the posterolateral approach. Am. J. Orthop. 35 (5), 218– 224. Brand, R.A., Pedersen, D.R., Davy, D.T., et al., 1994. Comparison of hip force calculations and measurements in the same patient. J. Arthroplasty 9, 45–51. Callaghan, J.J., Fulghum, C.S., Glisson, R.R., Stranne, S.K., 1992. The effect of femoral stem geometry on interface motion in uncemented porous-coated total hip prosthesis. J. Bone Joint Surg. 74, 839–848. Chung, W.K., Liu, D., Foo, L.S.S., 2004. Mini-incision total hip replacement-surgical technique and early results. J. Orthop. Surg. 12, 19–24. De Leva, P., 1996. Adjustments to zatsiorsky-seluyanov’s segment inertia parameters. J. Biomech. 29, 1223–1230.
134
D. Glaser et al. / Clinical Biomechanics 23 (2008) 127–134
Dennis, D.A., Komistek, R.D., Hoff, W.A., Gabriel, S., 1996. In vivo knee kinematics derived using an inverse perspective technique. Clin. Orthop. Relat. Res. 331, 107–117. Dennis, D.A., Komistek, R.D., Northcut, E.J., et al., 2001. In vivo determination of hip joint separation and the forces generated due to impact loading conditions. J. Biomech. 34, 623–629. DiGioia III, A.M., 2003. The role of navigational tools in more accurate and less invasive THA. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA, Symposium F. Fiedler, E., Hildebrand, M., Moisio, K.C., Wimmer, M.A., et al., 2006. MIS approach to THA affects gait differences. In: American Academy of Orthopaedic Surgeons AAOS Meeting, Chicago, IL. Harris, W.H., Sledge, C.B., 1990. Total hip and total knee replacement. N. Engl. J. Med., 323–725. Hewitt, J.D., Glisson, R.R., Guilak, A.K., et al., 2002. The mechanical properties of the human hip capsule ligaments. J. Arthroplasty 17, 82– 89. Howell, J.R., Masri, B.A., Duncan, C.P., 2004. Minimally invasive versus standard incision anterolateral hip replacement: a comparative study. Orthop. Clin. North. Am. 35, 153–162. Kane, T.R., Levinson, D.A., 1985. Dynamics: Theory and Applications. McGraw-Hill, NY. Kane, T.R., Levinson, D.A., 1996. Dynamics Online: Theory and Implementation with Autolev, Online Dynamics, Sunnyvale, CA. Komistek, R.D., Stiehl, J.B., Paxson, R.D., 1994. Mathematical model of the human lower extremity. In: Vossoughi, J. (Ed.), Biomedical Engineering Recent Developments, Washington DC, pp. 1098–1101. Komistek, R.D., Stiel, J.B., Dennis, D.A., et al., 1998. Mathematical model of the lower extremity joint reaction forces using Kane’s method of dynamics. J. Biomech. 31, 185–189. Komistek, R.D., Dennis, D.A., Ochoa, J.A., et al., 2002. In vivo comparison of hip separation after metal-on-metal or metal-on polyethylene total hip arthroplasty. J. Bone Joint Surg. 84A, 1836–1841. Lewinnek, G.E., Lewis, J.L., Tarr, R., et al., 1978. Incidence of dislocation following hip arthroplasty for patients in the rehabilitation setting. J. Bone Joint Surg. Am. 60, 217–220.
Mahfouz, M.R., Hoff, W.A., Komistek, R.D., Dennis, D.A., 2003. A robust method for registration of three-dimensional knee implant models to two-dimensional fluoroscopy images. IEEE Trans. Med. Imaging 22, 1561–1574. Mardones, R., Pagnano, M., Nemanich, J., et al., 2005. Muscle damage after total hip arthroplasty done with the two-incision and mini-posterior techniques. Clin. Orthop. Relat. Res. 441, 63– 67. Northcut, E.J., 1998. In vivo determination of hip joint separation. Colorado school of mines. Engineering Thesis. Pagnano, M.W., Leone, J., Meneghini, R.M., et al., 2007a. A prospective randomized trial shows two-incision THAs do not recover quicker than mini-posterior THAs. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Pagnano, M.W., Meneghini, R.M., Kaufman, K.R., et al., 2007b. No benefit of the two-incision THA over mini-posterior THA: a comprehensive gait analysis study. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Pavone, V., Chimemto, G., Sharrock, N., Sculco, T., 2001. The role of incision length in total hip arthroplasty. J. Bone Joint Surg. 83B (Suppl 2), 213S. Sarojak, M.E., 1998. Model-fit: an interactive pose determining system. Colorado school of mines. Engineering Thesis, Golden, CO. Sculco, T.P., 2003. Advances in minimally invasive surgery. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Sirianni, L.A., Dorr, L.D., Yun, A., et al., 2007. Gait analysis of MIS THR. In: American Academy of Orthopaedic Surgeons AAOS Meeting, San Diego, CA. Swanson, T.V., Hanna, R.S., 2003. Advantages of cementless THA using minimally invasive surgical technique. In: American Academy of Orthopaedic Surgeons AAOS Meeting, New Orleans, LA. Wenz, J.F., Gurkan, I., Jibodh, S.R., 2002. Mini-incision total hip arthroplasty: a comparative assessment of perioperative outcomes. Orthopedics 25, 1031–1043. Winter, D.A., 1990. Biomechanics and Motor Control of Human Movement. John Wiley & Sons, Inc., Canada.