Michael E Berend

Saint Francis Health System, Tulsa, Oklahoma, United States

Are you Michael E Berend?

Claim your profile

Publications (107)243.34 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: Osteoarthritis (OA) is characterized by pain, loss of function, joint space narrowing, and possible limb deformity. The main changes in knee as a result of OA occur at the articular cartilage, yet there are a variety of changes that can occur to the knee either preceding the development of OA or due to the degeneration. These include damage to soft tissues such as the ligaments and menisci, along with loss of the mechanoreceptors needed to maintain joint function, misalignment of the mechanical axes and resultant changes in stress, and altered weight distribution through the bottom of the foot.[1,2] These changes require different measurement tools to accurately assess the degree of OA severity. A single test that could help evaluate these changes would be a valuable addition to the field. Pressure sensors can be used to reliably and accurately determine a patient’s weight distribution throughout the bottom of the feet. The changes in weight distribution that can be determined with this tool have the potential to provide insight into the condition of the knee as a result of OA damage. The entire lower limb functions as a continuous system, and it is likely that variation in part of this system can be measured and related to changes in another part of it. The purpose of this study was to measure the plantar force distribution in patients with knee OA and to determine if there is any relationship to certain changes seen in the knee, including joint alignment and the extent of soft tissue damage. Methods: A total of 37 subjects (IRB #1139155) were enrolled in the alignment study; 23 of these subjects were included in the damage model. Weight Distribution: Weight distribution data was collected for each patient. Two 20-second unassisted, quiet standing trials were completed prior to surgery using the Matscan pressure mat (Tekscan, Boston, MA). To determine medial weight distribution, a template profile was placed around each foot and adjusted to best fit the areas of measured force, using the middle toe and heel as a line of demarcation between medial and lateral weight distribution. Alignment Models: Initial evaluation of the relationship between medial weight distribution and alignment was completed using data on knees that were operated on. A general linear model, assuming linearity, was constructed using generalized estimating equations (GEE) to examine the relationship between weight distribution and knee alignment on the population of surgery knees only. Repeated measures were used to account for subjects that had surgery on both knees. The model accounted for age, gender, BMI, history of surgical intervention, and which leg (or both) was operated on. SAS statistical software (SAS, North Carolina) was used and p < 0.05 denoted statistical significance. All subsequent evaluations used a complete data set of knees with OA. Following initial evaluation, a generalized linear model was constructed using GEE. Repeated measures were used to account for some of the subjects having two knees with OA. Damage Models: The relationships between weight distribution and various measures of knee damage (femoral/tibial chondromalacia, femoral/tibial osteophytes, meniscus damage, and ACL damage) were evaluated using ordinal logistic regression with repeated measures. A 95% CI was calculated for each odds ratio. Any confounders and other variables believed to play a role in explaining the response were included in the model, along with any interactions of interest. Unadjusted and adjusted models were also examined for each response. The models for chondromalacia, osteophytes, and meniscus damage were adjusted for age, sex, BMI, knee compartment (medial/lateral), history of surgical intervention, presence of OA in the compartment, and the leg/side of interest. The interaction between weight distribution and compartment was included to evaluate how weight distribution might affect the odds of damage for each compartment, separately. The model for ACL condition accounted for age, sex, BMI, history of surgical intervention, and side of interest. Results: Alignment models: After accounting for age, sex, BMI, which knee (or both) was operated on, and history of surgery on the lower limb, medial plantar weight distribution was found to have a significant effect on alignment (0.19, p=0.002). As the proportion of plantar weight transmitted through the medial side of the foot increased, the knee alignment angle increased and moved in a varus-tovalgus direction. Increasing BMI was also statistically significant (0.41, p < 0.001). This indicates that a higher BMI results in alignment changes from a varus-to-valgus direction. A positive history of lower limb surgery was significant in the model with an effect of decreasing the alignment angle and moving towards a varus alignment (-6.71, p < 0.001). Damage Models: Plantar medial weight distribution was significant in the damage model for ACL condition (-.14, p=0.029). This means that an increase in medial plantar weight distribution decreases the likelihood of having more severe levels of ACL damage. For all damage models in which it was considered, a diagnosis of OA was associated with increased likelihood of damage and was statistically significant (p < 0.001). Finally, being female resulted in a significantly greater likelihood of having more severe levels of damage (1.64, p=0.045). The odds ratio results confirmed the significant relationship between ACL damage and plantar medial weight distribution (OR = 0.872, CI [0.772, 0.986]). Discussion: A direct relationship between percentage of weight on the medial portion of the foot and the alignment of OA knees was observed. A positive knee angle corresponded with a valgus alignment in this study; therefore, increasing medial weight was related to moving towards a valgus (or more valgus) alignment. The vast majority of knees included in the alignment model were in varus alignment. This means that the increase in angle related to increasing medial weight was more indicative of a movement away from varus and towards a less severe varus angle (closer to a neutral alignment). This tendency to move away from a varus alignment as medial weight increases is beneficial when considering knee OA. A neutral angle is optimal to lower the effects that knee alignment can have on OA. With the known relationship between varus alignment and OA, a clinician might want to be more aware of possible OA or even suggest corrective measures that can provide for a more even plantar weight distribution, such as braces, orthotics, or physical therapy. The inverse relationship between medial plantar weight distribution and ACL damage indicates that as the percentage of plantar weight increases, the odds of observing more severe ACL condition decrease. Nearly all of the patients included in the damage model had uneven plantar weight distribution, with a lower portion of the total weight on the medial half of the foot. Therefore, the relationship observed can be thought of as the odds of increased ACL damage decreasing as the medial weight percentage increases and the weight becomes more evenly distributed on the bottom of the foot. Another interesting note from the ACL damage model was the effect of gender on the likelihood of damage. It is well established that women have an increased likelihood of ACL injuries compared to men, which reinforces the validity of the current study. Significance: OA is a damaging disease that can impact lower limb function. This study examined how plantar weight distribution was related to the changes in knee alignment and the various types of joint damage in patients with OA. Relationships were observed between medial plantar weight distribution and knee alignment angles, BMI, gender, and ACL damage. ORS 2015 Annual Meeting
    Orthopaedic Research Society, Las Vegas, NV; 03/2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: Osteoarthritis (OA) is characterized by pain, loss of function, joint space narrowing, and possible limb deformity. The main changes in knee as a result of OA occur at the articular cartilage, yet there are a variety of changes that can occur to the knee either preceding the development of OA or due to the degeneration. These include damage to soft tissues such as the ligaments and menisci, along with loss of the mechanoreceptors needed to maintain joint function, misalignment of the mechanical axes and resultant changes in stress, and altered weight distribution through the bottom of the foot.[1,2] These changes require different measurement tools to accurately assess the degree of OA severity. A single test that could help evaluate these changes would be a valuable addition to the field. Pressure sensors can be used to reliably and accurately determine a patient’s weight distribution throughout the bottom of the feet. The changes in weight distribution that can be determined with this tool have the potential to provide insight into the condition of the knee as a result of OA damage. The entire lower limb functions as a continuous system, and it is likely that variation in part of this system can be measured and related to changes in another part of it. The purpose of this study was to measure the plantar force distribution in patients with knee OA and to determine if there is any relationship to certain changes seen in the knee, including joint alignment and the extent of soft tissue damage. Methods: A total of 37 subjects (IRB #1139155) were enrolled in the alignment study; 23 of these subjects were included in the damage model. Weight Distribution: Weight distribution data was collected for each patient. Two 20-second unassisted, quiet standing trials were completed prior to surgery using the Matscan pressure mat (Tekscan, Boston, MA). To determine medial weight distribution, a template profile was placed around each foot and adjusted to best fit the areas of measured force, using the middle toe and heel as a line of demarcation between medial and lateral weight distribution. Alignment Models: Initial evaluation of the relationship between medial weight distribution and alignment was completed using data on knees that were operated on. A general linear model, assuming linearity, was constructed using generalized estimating equations (GEE) to examine the relationship between weight distribution and knee alignment on the population of surgery knees only. Repeated measures were used to account for subjects that had surgery on both knees. The model accounted for age, gender, BMI, history of surgical intervention, and which leg (or both) was operated on. SAS statistical software (SAS, North Carolina) was used and p < 0.05 denoted statistical significance. All subsequent evaluations used a complete data set of knees with OA. Following initial evaluation, a generalized linear model was constructed using GEE. Repeated measures were used to account for some of the subjects having two knees with OA. Damage Models: The relationships between weight distribution and various measures of knee damage (femoral/tibial chondromalacia, femoral/tibial osteophytes, meniscus damage, and ACL damage) were evaluated using ordinal logistic regression with repeated measures. A 95% CI was calculated for each odds ratio. Any confounders and other variables believed to play a role in explaining the response were included in the model, along with any interactions of interest. Unadjusted and adjusted models were also examined for each response. The models for chondromalacia, osteophytes, and meniscus damage were adjusted for age, sex, BMI, knee compartment (medial/lateral), history of surgical intervention, presence of OA in the compartment, and the leg/side of interest. The interaction between weight distribution and compartment was included to evaluate how weight distribution might affect the odds of damage for each compartment, separately. The model for ACL condition accounted for age, sex, BMI, history of surgical intervention, and side of interest. Results: Alignment models: After accounting for age, sex, BMI, which knee (or both) was operated on, and history of surgery on the lower limb, medial plantar weight distribution was found to have a significant effect on alignment (0.19, p=0.002). As the proportion of plantar weight transmitted through the medial side of the foot increased, the knee alignment angle increased and moved in a varus-tovalgus direction. Increasing BMI was also statistically significant (0.41, p < 0.001). This indicates that a higher BMI results in alignment changes from a varus-to-valgus direction. A positive history of lower limb surgery was significant in the model with an effect of decreasing the alignment angle and moving towards a varus alignment (-6.71, p < 0.001). Damage Models: Plantar medial weight distribution was significant in the damage model for ACL condition (-.14, p=0.029). This means that an increase in medial plantar weight distribution decreases the likelihood of having more severe levels of ACL damage. For all damage models in which it was considered, a diagnosis of OA was associated with increased likelihood of damage and was statistically significant (p < 0.001). Finally, being female resulted in a significantly greater likelihood of having more severe levels of damage (1.64, p=0.045). The odds ratio results confirmed the significant relationship between ACL damage and plantar medial weight distribution (OR = 0.872, CI [0.772, 0.986]). Discussion: A direct relationship between percentage of weight on the medial portion of the foot and the alignment of OA knees was observed. A positive knee angle corresponded with a valgus alignment in this study; therefore, increasing medial weight was related to moving towards a valgus (or more valgus) alignment. The vast majority of knees included in the alignment model were in varus alignment. This means that the increase in angle related to increasing medial weight was more indicative of a movement away from varus and towards a less severe varus angle (closer to a neutral alignment). This tendency to move away from a varus alignment as medial weight increases is beneficial when considering knee OA. A neutral angle is optimal to lower the effects that knee alignment can have on OA. With the known relationship between varus alignment and OA, a clinician might want to be more aware of possible OA or even suggest corrective measures that can provide for a more even plantar weight distribution, such as braces, orthotics, or physical therapy. The inverse relationship between medial plantar weight distribution and ACL damage indicates that as the percentage of plantar weight increases, the odds of observing more severe ACL condition decrease. Nearly all of the patients included in the damage model had uneven plantar weight distribution, with a lower portion of the total weight on the medial half of the foot. Therefore, the relationship observed can be thought of as the odds of increased ACL damage decreasing as the medial weight percentage increases and the weight becomes more evenly distributed on the bottom of the foot. Another interesting note from the ACL damage model was the effect of gender on the likelihood of damage. It is well established that women have an increased likelihood of ACL injuries compared to men, which reinforces the validity of the current study. Significance: OA is a damaging disease that can impact lower limb function. This study examined how plantar weight distribution was related to the changes in knee alignment and the various types of joint damage in patients with OA. Relationships were observed between medial plantar weight distribution and knee alignment angles, BMI, gender, and ACL damage.
    Orthopaedic Research Society, Las Vegas, NV; 03/2015
  • Open Journal of Orthopedics 01/2015; 05(05):115-119. DOI:10.4236/ojo.2015.55015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Return to sexual activity is important to patients, but there is limited information regarding sexual function following total hip arthroplasty (THA) and total knee arthroplasty (TKA). A multicenter study of 806 THA, 542 TKA, and 181 control patients less than 60 years of age was conducted using an independent survey center to question subjects about their sexual function. Only 1.3% of THA and 1.6% of TKA patients stated they were not sexually active due to their operation. No statistically significant differences were noted in any sexual function outcome categories based on the bearing surface, femoral head size, or use of surface replacement arthroplasty in the hip cohort. Multivariate analysis revealed no difference in the percentage of patients sexually active following a THA or TKA (OR 1.19, p = 0.38). Most young active patients return to their baseline or higher level of sexual activity after hip and knee arthroplasty.
    The Journal of Arthroplasty 10/2014; 30(2). DOI:10.1016/j.arth.2014.09.029 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Uncemented stems have been used in THA for well over two decades, but there are relatively few studies reporting on the results after 20 years.
    Clinical Orthopaedics and Related Research 07/2014; 473(2). DOI:10.1007/s11999-014-3763-y · 2.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study was to investigate the use for screws and cement, and primary and revision specific prosthesis for revision TKR. Between July 1989 and February 2010, 839 consecutive revision TKAs were performed, with 609 knees meeting inclusion criteria. At 17 years followup, Kaplan-Meier survivorship was .9859 for revision specific prosthesis with screws and cement, .9848 for revision prosthesis with no screws, 0.9118 for primary prosthesis with screws, and .9424 for primary prosthesis with no screws. Revision TKRs using screws had greater defects (p < .0001). Use of revision prosthesis along with screws and cement to correct largely defective revision TKRs is highly recommended.
    The Journal of Arthroplasty 01/2014; DOI:10.1016/j.arth.2014.07.027 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study was to investigate screws and cement for large tibial bone defects during primary TKA. Of 14,686 consecutive primary TKAs performed between December 1988 and February 2010, 256 received screws and cement for tibial defects. Cox regression was used for the analysis. 20-year survival probability was 0.9897 (screws) and 0.9339 (no screws) (P=.4225 log-rank). Tibial bone condition was significantly worse in knees receiving screws (P<.0001) with 73.0% having defects in the screws group and 3.4% (P<.0001) for non-screws. Radiolucency appeared in 13.7% (screws) and 6.4% (no screws) postoperatively. Screws were $137 each, wedges $910 to $2240. Knees with tibial defects and screws performed similarly if not better than knees without defects at substantially lower cost than alternatives.
    The Journal of arthroplasty 12/2013; 29(6). DOI:10.1016/j.arth.2013.12.023 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Limited experimental data exist comparing the mechanical response of the tibial cortex between fixed and rotating platform (RP) total knee arthroplasty (TKA), particularly in the revision setting. We asked if RP-TKA significantly affects tibiofemoral torque and cortical stain response in both the primary and revision settings. Fixed and RP tibial trays were implanted into analogue tibias and biomechanically tested under axial and torsional loading. Torque and strain response were analyzed using digital image correlation. Fixed bearing designs exhibited 13.8 times greater torque (P<0.01), and 69% (P<0.01) higher cortical strain than RP designs. Strain response was similar in the primary and revision cohorts. The decrease in torque transfer could act as a safeguard to reduce stress, micromotion and torsional fatigue in scenario of poor bone stock.
    The Journal of arthroplasty 11/2013; 29(3). DOI:10.1016/j.arth.2013.08.024 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Revision knee data from six joint arthroplasty centers were compiled for 2010 and 2011 to determine mechanism of failure and time to failure. Aseptic loosening was the predominant mechanism of failure (31.2%), followed by instability (18.7%), infection (16.2%), polyethylene wear (10.0%), arthrofibrosis (6.9%), and malalignment (6.6%). Mean time to failure was 5.9years (range 10 days to 31years). 35.3% of all revisions occurred less than 2years after the index arthroplasty, 60.2% in the first 5years. In contrast to previous reports, polyethylene wear is not a leading failure mechanism and rarely presents before 15years. Implant performance is not a predominant factor of knee failure. Early failure mechanisms are primarily surgeon-dependent.
    The Journal of arthroplasty 08/2013; 28(8). DOI:10.1016/j.arth.2013.04.056 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Many previous biomechanical studies of bone and bone substitutes have estimated strains in these materials using strain gages. The purpose of this study was to compare digital image correlation (DIC) strain measurements to those obtained from strain gages in order to assess the applicability of DIC technology to common biomechanical testing scenarios. Compression and bending tests were conducted on aluminum alloy, polyurethane foam, and laminated polyurethane foam specimens. Results showed no significant differences in the principal strain values (or the variances) between strain gage and DIC measurements on the aluminum alloy and laminated polyurethane foam specimens. There were significance differences between the principal strain measurements of the non-laminated polyurethane foam specimens, but the deviation from the theoretical results was similar for both measurement techniques. In summary, DIC techniques provide similar results to those obtained from strain gages and also provide full field strain results.
    ASME 2013 Summer Bioengineering Conference; 06/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Potential advantages suggested but not confirmed for surface replacement arthroplasty (SRA) over THA include lower frequency of limp, less thigh pain, less limb length discrepancy, and higher activity. QUESTIONS/PURPOSES: We therefore determined whether patients having SRA had a limp, thigh pain, or limb length discrepancy less frequently or had activity levels higher than patients having THA. METHODS: In a multicenter study, we surveyed 806 patients aged 18 to 60 years with a premorbid UCLA activity score of 6 or more who underwent hip arthroplasty for noninflammatory arthritis at one of five orthopaedic centers. Patients had either a cementless THA with an advanced bearing surface (n = 682) or an SRA (n = 124). The patients were demographically comparable. Specific telephone survey instruments were designed to assess limp, thigh pain, perception of limb length, and activity levels. Minimum followup was 1 year (mean, 2.3 years; range, 1.1-3.9 years). RESULTS: When controlled for age, sex, and premorbid activity level, patients with SRA had a higher incidence of complete absence of any limp, lower incidence of thigh pain, lower incidence of perception of limb length discrepancy, greater ability to walk continuously for more than 60 minutes, higher percentage of patients who ran after surgery, greater distance run, and higher percentage of patients who returned to their most favored recreational activity. CONCLUSIONS: When interviewed by an independent third party, patients with SRA reported higher levels of function with fewer symptoms and less perception of limb length discrepancy compared to a similar cohort of young, active patients with THA. LEVEL OF EVIDENCE: Level III, therapeutic study. See Instructions for Authors for a complete description of levels of evidence.
    Clinical Orthopaedics and Related Research 03/2013; 471(12). DOI:10.1007/s11999-013-2915-9 · 2.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: With increased precision in alignment offered by new generations of instrumentation and customized guides, this study was designed to establish a biomechanically-based target alignment for the balance of tibial loading in order to diminish the likelihood of pain and subsidence related to mechanical overload post-UKA. Sixty composite tibias were implanted with Oxford UKA tibial components with varied sagittal slope, resection depth, rotation and medial shift using patient matched instrumentation. Digital image correlation and strain gage analysis was conducted in static loading to evaluate strain distribution as a result of component alignment. In this model, minimal distal resection and most lateral positioning, neutral component rotation, and 3°of slope (from mechanical axis) exhibited the most balanced strain response to loading following UKA.
    The Journal of arthroplasty 03/2013; 28(9). DOI:10.1016/j.arth.2013.01.004 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Introduction Mobile bearing total knee arthroplasty (TKA) tibial components can allow for high tibiofemoral conformity while minimizing polyethylene contact stress and reducing bone-implant interface stress. Prior studies have investigated the cortical strain variance between fixed and mobile bearing components in a limited number of measurement regions. However, no current experimental data exists across the entire proximal tibial cortex, and no comparisons have been made within the context of knee arthroplasty revision surgery. The purpose of this study was to investigate the influence of bearing mobility on torque and torsional strain across the entire cortical surface of the proximal tibia in the primary and revision setting. Specifically of interest is the change in induced strain, the instance of femoral component rotation and rotational malalignment. Methods In order to compare the mechanical response of the tibia following implantation of fixed and rotating platform mobile-bearing TKA components, four experimental groups were included in this study: 1) Fixed-bearing, posterior stabilized primary components (PFC Sigma, DePuy, Warsaw, IN); 2) Rotating platform posterior stabilized primary components (PFC Sigma, DePuy, Warsaw, IN); 3) Fixed-bearing posterior stabilized revision components with 115 mm press-fit distal stem (PFC TC3, DePuy, Warsaw, IN; 4) Rotating platform posterior stabilized revision components with 75 mm press-fit distal stem (PFC TC3, DePuy, Warsaw, IN). Six components in each experimental group were implanted into fourth generation composite tibia specimens (Pacific Research Laboratories, Vashon, WA) using proximal cementing and standard instrumentation. Following implantation, tibias were prepared for surface strain quantification through the use of complimentary strain gage and digital image correlation (DIC) methodologies. Two three-element rectangular rosette strain gages were applied to the tibia in the anteromedial and posterolateral quadrants for direct strain measurement. Digital image correlation techniques (Aramis 6.0, Gom, Inc., Braunschweig, Germany) were used to obtain full-field strain measurements 360 degrees around the proximal tibial cortex. For DIC measurement, a black and white speckled paint was applied to the tibia surface, which was then optically tracked with a set of two high definition cameras throughout the loading cycle. To obtain full-field strain measurements around the entire tibia, tests were repeated in four independent viewing angles. Deformation and strain measurements were calculated within the DIC system software and then analyzed utilizing a custom merging algorithm (Matlab R2012a, Mathworks, Natick, MA) to combine data from specimens and repeated trials within each experimental group. Biomechanical testing was conducted on a biaxial electrodynamic materials testing machine (E10,000 A/T, Instron, Norwood, MA). Specimens were incorporated into the materials testing machine via a custom fixture allowing free x-y translation of the potted base. Appropriate femoral components were integrated into the upper testing grips to allow for repeatable load application through the femoral component onto the polyethylene bearing surface. A silicon-based lubricant (DM-Fluid-350CS, ShinEtsu Chemical Co, Tokyo) was applied between all articulating surfaces to mimic in vivo frictional characteristics. Testing was conducted in two phases: 1) Compressive loading followed by a 5 degree internal rotation with femoral component in a full extension, and 2) Compressive loading followed by a 10 degree external rotation with the femoral component 90 degrees of flexion. In both instances the tibia was loaded at a rate of 60 N/s to a peak load of 2.5 kN, while femoral component rotation was introduced at a rate of 0.5 ˚/s. Five trials were repeated for each of the four DIC viewing angles in all 24 specimens. Statistical analysis was performed utilizing paired t-tests to evaluate significant differences between designs in torque response. Further examination was conducted to evaluate contribution of torsional strain to the total overall strain response in each DIC measurement region. Statistical significance was indicated at p ≤ 0.05. Results Torsional Response: Average torsional moments during rotational testing in both 0 and 90 degrees of flexion are presented in Table 1. In the primary setting, fixed bearing tibias generated 13.7 times the torsional moment of the rotating platform primary design (p<0.01) when the extended femoral component was rotated 5 degrees internally. Torsional moments in the fixed tibias were 11.2 times greater than those in the rotating platform designs in the revision setting (p<0.01). In flexion, fixed bearing designs generated 4.4 times greater torque in the primary (p<0.01) and 4.8 times greater torque in the revision setting (p<0.01) when the femoral component was rotated 10 degrees externally. Strain Response: Representative anterior and posterior DIC strain responses to compressive and torsional loading are presented in Figure 1. Torsional strain response was seen to diminish substantially when rotating platform devices were utilized, most notably in the posterior tibia. This diminished strain response to torsional loading was consistent in both primary and revision tibial trays. In order to numerically quantify DIC data, each field of view was divided into 5 measurement regions, in order from most proximal to most distal, for von Mises strain averaging (Figure 1). Average cortical strain for each measurement region was calculated by merging 8,000 – 30,000 individual strain data points collected within the given region throughout five repeated trials of six specimens in each respective experimental group. As a subset of all DIC strain analysis, the von Mises strain data for fixed and rotating platform primary knee components, with 10˚ external femoral rotation is presented in Table 2. In the primary fixed bearing components, average cortical strain in 6 of 10 measurement regions significantly increased between 17% (p=.0004) and 56% (p=.0001) when the femoral component was rotated 10˚ externally. Conversely, there was no statistically significant change in strain induced in the primary rotating platform group with the introduction of external femoral rotation. In the fixed bearing revision setting (not presented in Table 2), a significant increase from 18% (p=.0001) to 69% (p=.0001) increase in strain was observed with 10˚ of external femoral component rotation in 6 of 10 measurement regions. Similar to the primary components, there was no statistically significant change in strain due to femoral rotation in any anterior or posterior measurement regions in the rotating platform design. Discussion Femoral component rotation about the tibial tray occurs cyclically during the gait cycle and generates a torsional moment at the tibial tray. Relative femoral component rotation and subsequent torsional moments can also be generated in the event of suboptimal rotational positioning during implantation. Both the primary and revision rotating platform designs exhibited vastly reduced torque response between the articulating femoral and tibial components when under compressive loading and femoral rotation. Strain response in the tibia was significantly altered in the majority of measurement regions when the femoral component was rotated on the tibial tray in the fixed bearing designs. However, no significant change in cortical strain loading was observed in anterior and posterior regions when the femoral component was rotated in the rotating platform designs. The clinical application of this study may be limited due to the use of composite, rather than cadaveric tibial specimens. Furthermore, no muscular or ligamentous forces were replicated during loading. Nevertheless, this model is effective in the direct comparison between strain and torsional response fixed and mobile-bearing designs. Significance In a comparison between fixed and rotating platform tibial component designs in both the primary and revision setting, rotating platform tibial trays demonstrated significantly less transfer of rotational moment between the femoral and tibial component than their fixed-bearing counterparts. As a result of this diminished torsional load transfer, minimal increases in cortical strains were observed during femoral component rotation in the rotating platform study group. The decrease in torque transfer between tibial tray and implanted bone in mobile-bearing technology may act as a safeguard to reduce stress and torsional fatigue at the bone-implant interface.
    Orthopaedic Research Society, San Antonio, TX; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Implant survival after total knee arthroplasty has historically been dependent on postoperative knee alignment, although failure may occur when alignment is correct. Preoperative knee alignment has not been thoroughly evaluated as a possible risk factor for implant failure after arthroplasty. The purpose of this study was to analyze the effect of preoperative knee alignment on implant survival after total knee arthroplasty. We performed a retrospective review of 5342 total knee arthroplasties performed with use of cemented Anatomic Graduated Component implants from 1983 to 2006. Each knee was independently measured preoperatively and postoperatively for overall coronal alignment. Neutral ranges for preoperative and postoperative alignment were defined by means of Cox proportional hazards regression. The overall failure rate was 1.0% (fifty-four of 5342 prostheses); failure was defined as aseptic loosening of the femoral and/or tibial component. The average preoperative anatomical alignment (and standard deviation) was 0.1° ± 7.7° of varus (range, 25° of varus to 35° of valgus), and the average postoperative anatomical alignment (and standard deviation) was 4.7° ± 2.5° of valgus (range, 12° of varus to 20° of valgus). The failure rate in knees in >8° of varus preoperatively (2.2%; p = 0.0005) or >11° of valgus preoperatively (2.4%; p = 0.0081) was elevated when compared with knees in neutral preoperatively (0.71%). Knees with preoperative deformities corrected to postoperative neutral alignment (2.5° through 7.4°) had a lower failure rate (1.9%) than undercorrected or overcorrected knees (3.0%) (p = 0.0103). Knees with postoperative neutral alignment, regardless of preoperative alignment, had a lower failure rate (0.74%) than knees with postoperative alignment of <2.5° or >7.4° of anatomic valgus (1.7%) (p < 0.0001). Patients with excessive preoperative alignment (>8° of varus or >11° of valgus) have a greater risk of failure (2.3%). Neutral postoperative alignment (2.5° through 7.4° of valgus) improves (1.9% for preoperatively deformed knees) but does not completely eliminate the risk of failure (0.5% for knees that were neutral both preoperatively and postoperatively). Careful attention should be paid to knee alignment during total knee arthroplasty, especially for patients with severe preoperative deformities. Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.
    The Journal of Bone and Joint Surgery 01/2013; 95(2):126-31. DOI:10.2106/JBJS.K.00607 · 4.31 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Initial stability with limited micromotion in uncemented total hip arthroplasty acetabular components is essential for bony attachment and long-term biomechanical fixation. This study compared porous titanium fixation surfaces to clinically established, plasma-sprayed designs in terms of interface stability and required seating force. Porous plasma-sprayed modular and metal-on-metal (MOM) cups were compared to a modular, porous titanium designs. Cups were implanted into polyurethane blocks with1-mm interference fit and subsequently edge loaded to failure. Porous titanium cups exhibited 23% to 65% improvement in initial stability when compared to plasma-sprayed cup designs (P=.01): a clinically significant increase, based on experience and prior literature. The results of this study indicate increased interface stability in porous titanium-coated cups without significantly increasing the necessary force and energy required for full seating.
    The Journal of arthroplasty 11/2012; 28(3). DOI:10.1016/j.arth.2012.07.035 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Press-fit acetabular components are susceptible to deformation in an underreamed socket, with excessive deformation of metal-on-metal (MOM) components potentially leading to increased torsional friction and micromotion. Specifically, however, it remains unclear how cup diameter, design, and time from implantation affect shell deformation. QUESTIONS/PURPOSES : We asked whether (1) changes in component geometry and material altered maximum shell deformation and (2) time-dependent deformational relaxation processes occurred. METHODS: Diametral deformation was quantified after press-fit implantation of metal shells into a previously validated polyurethane model. Experimental groups (n = 6-8) consisted of 48-, 54-, 60-, and 66-mm MOM cups of 6-mm wall thickness, 58-mm cups of 10-mm wall thickness, and CoCrMo and Ti6Al4V 58-mm modular cups. RESULTS : Greater cup diameter, thinner wall construction, and Ti6Al4V modular designs generated conditions for maximum shell deformation ranging from 0.047 to 0.267 mm. Relaxation (18%-32%) was observed 120 hours postimplantation in thin-walled and modular designs. CONCLUSIONS : Our findings demonstrate a reduction of shell deformation over time and suggest, under physiologic loading, early component deformation varies with design. CLINICAL RELEVANCE : Component deformation should be a design consideration regardless of bearing surface. Designs neglecting to adequately address deformational changes in vivo could be susceptible to diminished cup survival, increased wear, and premature revision.
    Clinical Orthopaedics and Related Research 09/2012; 471(2). DOI:10.1007/s11999-012-2553-7 · 2.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this multicenter study is to determine the utility of the erythrocyte sedimentation rate, C-reactive protein, and synovial fluid white blood cell (WBC) count and differential for evaluating periprosthetic joint infection (PJI) in patients with a failed unicompartmental knee arthroplasty (UKA). A total of 259 patients undergoing revision of a failed UKA were reviewed; 28 (10.8%) met the study criteria for PJI. The optimal cutoff values were 27 mm/h for the erythrocyte sedimentation rate, 14 mg/L for the C-reactive protein, 6200/mu L for the synovial fluid WBC count, and 60% for the differential. These tests are useful for diagnosing PJI after UKA with optimal cutoff values that are similar to those used for total knee arthroplasty; however, the optimal synovial WBC count was found to be somewhat higher, which may be related to the unresurfaced compartments. In addition, we found that nearly half of patients had suboptimal evaluation for PJI.
    The Journal of Arthroplasty 09/2012; 27(8):46-50. DOI:10.1016/j.arth.2012.03.033 · 2.37 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Acetabular cup orientation has been shown to influence dislocation, impingement, edge loading, contact stress, and polyethylene wear in total hip arthroplasty. Acetabular implant stiffness has been suggested as a factor in pelvic stress shielding and osseous integration. This study was designed to examine the combined effects of acetabular cup orientation and stiffness and on pelvic osseous loading. Four implant designs of varying stiffness were implanted into a composite hemipelvis in 35° or 50° of abduction. Specimens were dynamically loaded to simulate gait and pelvic strains were quantified with a grid of rosette strain gages and digital image correlation techniques. Changes in the joint reaction force orientation significantly altered mean acetabular bone strain values up to 67%. Increased cup abduction resulted in a 12% increase along the medial acetabular wall and an 18% decrease in strain in inferior lateral regions. Imbalanced loading distributions were observed with the stiffer components, resulting in higher, more variable, and localized surface strains. This study illustrates the effects of cup stiffness, gait, and implant orientation on loading distributions across the implanted pelvis.
    The Journal of arthroplasty 07/2012; 28(2). DOI:10.1016/j.arth.2012.05.026 · 2.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: INTRODUCTION: While long-term success in primary total knee arthroplasty (TKA) is high, aseptic loosening, subsidence, and tibial collapse remain as factors in tibial component revision. Prosthetic alignment, patient characteristics and implant design are all important factors in long-term survival of TKA, yet the level at which each of these factors contribute to implant loosening has not been fully described. The purpose of this study was to investigate the influence of bone resection depth on proximal tibial loading. METHODS: Three-dimensional models of 3rd generation composite tibiae (Medium, Left Tibia, Model 2150, Pacific Research Laboratories, Inc., Vashon, WA) were generated from CT scans using MIMICS medical imaging software (Version 13.0, Materialise, Belgium). All tibia models consisted of homogenous, isotropic material properties for each of the two layers (Ecortical = 7.6 GPa, Ecancellous = 104 MPa) within the composite tibia as specified by the manufacturer. The mesh and material property assignments were exported to ANSYS (Version 11.0, Canonsburg, PA) for analysis. Convergence tests and validation experiments were conducted to ensure the accuracy and validity of the models. Bone resection models were generated with 5- and 15-mm of resection beneath the joint line with 0 degrees of posterior slope. Depths of resection were chosen to reflect resection depth extremes based on knee deformity and bone quality observed clinically and are consistent with resection levels examined in previous studies. The resected tibia models were implanted with metal-backed TKA tibial components (Biomet Inc., Warsaw, IN). Based on surgeon guidance, size 75-mm tibial components were implanted into models with 5-mm of bone resection, which were matched with 70-mm femoral components for loading. Size 63-mm tibial components were implanted into models with 15-mm of bone resection and matched with 60-mm femoral components. Polyethylene bearing inserts of 10-mm thickness were matched with metal trays. All implant materials were modeled as homogeneous, isotropic materials (ECoCr = 230 GPa, EPE = 500 MPa, EPMMA = 2.28 GPa). The femoral component was positioned at 0° of flexion, with model contact areas verified through experimental contact analysis between manufactured femoral and tibial bearing components. The implanted tibia models were analyzed under axial compression. Static loads of 2700 N were applied through the femoral component in a balanced 50:50 (medial:lateral) distribution and a medially skewed 80:20 distribution to simulate varus loading. The distal 150-mm of the tibia was constrained against motion in all directions. RESULTS: To evaluate tibial loading, the strains were evaluated in the proximal 3-cm of the tibia, where tibial collapse is most likely. Furthermore, the tibia was circumferentially divided into 24 measurement regions. The delineated measurement regions are shown in Figure 1. In balanced 50:50 loading, the highest strain values were observed in the posterior regions of the tibia regardless of resection depth (Figure 2). Increased resection depth resulted in increased strain in all posterior measurement regions. It was noted that a 5-mm resection depth resulted in a smaller range of strains across all regions (1600 µε vs 2500 µε), i.e. the 5-mm resection depth had a more uniform strain distribution. In simulated varus loading, the highest strains were also observed in the posterior region and similar trends were observed for anterior and posterior behavior as compared to the balanced loading scenario. The 5-mm resection depth still resulted in a smaller range of strains (1800 µε vs 2800 µε). Strains for the AMP, AMC, and ALC regions were much greater than the strains for the balanced loading. The strains in the ALP regions for the valgus loading were approximately one-half the strains for the balanced load. An increased resection depth also resulted in increased posterior strains in all regions, with the PLP regions experiencing the greatest increase in strain. Within the PLP regions, the strain in the 0-1 cm region (i.e. on the periphery) increased approximately 200% when compared to the 5-mm resection depth for the same varus loading. DISCUSSION: Significantly higher strains were observed across a majority of the posterior tibia measurement regions for both loading scenarios as the resection depth increased. Peripheral anterior strains increased with an increased resection depth. This posterior and peripheral shift may be correlated to both the greater medial relative to lateral reduction of tibial plateau surface area with increased resection depth, in addition to the posterior shift in component stem placement required for implantation in a 15-mm resected tibia. Despite the experimental validation and clinical correlation of the results of this study, there are limitations to the present study. Composite tibiae have been used extensively in biomechanical studies, but it would be advantageous to consider a cadaveric specimen. The results of this study were validated with previous photoelastic coating experiments, but the model would be strengthened with comparison to strain gauge data. Inclusion of additional resection depths may also provide insight into the research question. SIGNIFICANCE: Prior clinical and biomechanical studies have indicated tibial overload as a cause of early TKA revision. Increased tibial resection depth may contribute to increased posterior and peripheral strain, particularly in the case of increased varus loading, in an area most susceptible to bony failure and early aseptic loosening
    Orthopaedic Research Society, San Francisco, CA; 02/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Partial knee arthroplasty has enjoyed renewed interest during the past decade. It is helpful to be familiar with the classic and current indications, contraindications, and technical aspects of partial knee arthroplasty, including patellofemoral, medial unicompartmental, and lateral unicompartmental knee arthroplasty. Various implant choices for partial knee arthroplasty can be compared and evaluated based on patient characteristics, design qualities, and reported outcomes. It is also helpful to review the indications and techniques for performing medial or lateral unicompartmental knee arthroplasty in combination with arthroscopically assisted reconstruction of the anterior cruciate ligament.
    Instructional course lectures 01/2012; 61:347-81.

Publication Stats

2k Citations
243.34 Total Impact Points

Institutions

  • 2014
    • Saint Francis Health System
      Tulsa, Oklahoma, United States
  • 2013
    • Spokane Joint Replacement Center
      Spokane, Washington, United States
    • Duke University Medical Center
      • Department of Orthopaedic Surgery
      Durham, North Carolina, United States
  • 2011–2013
    • Hip Knee Arkansas Foundation
      Little Rock, Arkansas, United States
  • 2002–2013
    • Saint Francis Hospital
      Tulsa, Oklahoma, United States
  • 2008
    • University of Missouri
      Columbia, Missouri, United States
  • 2007
    • St. Vincent Hospital
      Green Bay, Wisconsin, United States
  • 2006
    • University of Toledo
      Toledo, Ohio, United States