The use of technology to assess balance and alignment during total knee surgery can provide an overload of numerical data to the surgeon. Meanwhile, this quantification holds the potential to clarify and guide the surgeon through the surgical decision process when selecting the appropriate bone recut or soft tissue adjustment when balancing a total knee. Therefore, this paper evaluates the potential of deploying supervised machine learning (ML) models to select a surgical correction based on patient-specific intra-operative assessments. Based on a clinical series of 479 primary total knees and 1,305 associated surgical decisions, various ML models were developed. These models identified the indicated surgical decision based on available, intra-operative alignment, and tibiofemoral load data.Aims
Methods
Osteophytes in the posterior compartment of the knee pose a challenge in achieving soft tissue balance during total knee arthroplasty (TKA). Previous investigations have demonstrated the importance of various factors involved in obtaining flexion and extension gap balance, including the precision of femoral and tibial bone cuts as well as tensioning of the supporting pericapsular soft tissue structures (ligaments, capsule, etc.). However, the role of posterior compartment osteophytes has not been well studied. We hypothesize that space-occupying posterior structures affect soft tissue balance, especially in lesser degrees of flexion, in a cadaveric TKA model. Five cadaveric limbs were acquired. CT scans were obtained of each specimen to define the osseous contours. 3D printed specimen-specific synthetic osteophytes were fabricated in two sizes (10mm and 15mm). Posterior-stabilized TKAs were performed. Medial and lateral contact forces were measured during a passive range of motion using OrthoSensor ® (Dania Beach, FL) technology. For each specimen, trials were completed without osteophytes, and with 10mm and 15mm osteophytes applied to the posterior medial femur, with iterations at 0°, 10°, 30°, 45°, 60°, and 90° of flexion. These were recorded across each specimen in each condition for three trials. Tukey post hoc tests were used with a repeated measures ANOVA for statistical data analysis.Background
Methods
Sensor technology is seeing increased utility in joint arthroplasty, guiding surgeons in assessing the soft tissue envelope intra-operatively (OrthoSensor, FL, USA). Meanwhile, surgical navigation systems are also transforming, with the recent introduction of inertial measurement unit (IMU) based systems no longer requiring optical trackers and infrared camera systems in the operating room (i.e. OrthAlign, CA, USA). Both approaches have now been combined by embedding an IMU into an intercompartmental load sensor. As a result, the alignment of the tibial varus/valgus cut is now measured concurrently with the mediolateral tibiofemoral contact load magnitudes and locations. The wireless sensor is geometrically identical to the tibial insert trial and is placed on the tibial cutting plane after completing the proximal tibial cut. Subsequently, the knee is moved through a simple calibration maneuver, rotating the tibia around the heel. As a result, the sensor provides a direct assessment of the obtained tibial varus/valgus alignment. This study presents the validation of this measurement. In an in-vitro setting, sensor-based alignment measurements were repeated for several simulated conditions. First, the tibia was cut in near-neutral alignment as guided by a traditional, marker-based surgical navigation system (Stryker, MI, USA). Subsequently, the sensor was inserted and a minimum of five repeated sensor measurements were performed. Following these measurements, a 3D printed shim was inserted between the sensor and the tibial cutting plane, introducing an additional 2 or 4 degrees of varus or valgus, with the measurements then being repeated. Again, for each condition, a minimum of five sensor measurements were performed. Following completion of the tests, a computed tomography (CT) scan of the tibia was obtained and reconstructed using open source software (3DSlicer).Introduction and Aims
Method
The traditional method of soft-tissue balancing during TKA is subjective in nature, and stiffness and instability are common indications for revision, suggesting that TKA balancing by subjective assessment is suboptimal. This study examines the intraoperative mediolateral loads measured with a nanosensor-enabled tibial insert trial and the sequential balancing steps used to achieve quantitative balance. Data obtained from a prospective multicenter study was assessed to determine the effect of targeted ligament release on intra-articular loading, and to understand which types of releases are necessary to achieve quantified ligament balance. A group of 129 patients received sensor-assisted TKA, as part of a prospective multicenter study. Medial and lateral loading data were collected pre-release, during any sequential releases, and post-release. All data were collected at 10, 45, and 90 degrees during range of motion testing. Ligament release type, release technique type, and resultant loading were collected.Introduction & Aims
Methods
Substance abuse and dependence is thought to have a strongly negative impact on surgical outcomes. The purpose of this study was to determine the effects of drug misuse on total knee arthroplasty (TKA) revision incidence, revision causes, and time to revision by analyzing the Medicare database between 2005–2012. A retrospective review of the Medicare database within the PearlDiver Supercomputer (Warsaw, IN) for TKA and revisions was performed utilizing Current Procedural Terminology (CPT) and International Classification of Disease (ICD) ninth revision codes. Drug misuse was subdivided into cocaine, cannabis, opioids, sedative/hypnotic/anxiolytic, amphetamines, and alcohol. Time to revision, age, and gender were also investigated.Background
Methods
Successful clinical outcomes following unicompartmental knee arthroplasty (UKA) depend on component positioning, soft tissue balance and lower limb alignment, all of which can be difficult to achieve using manual instrumentation. A new robotic-guided technology has been shown to improve postoperative implant positioning and lower limb alignment in UKA but so far no studies have reported clinical results of robotic-assisted medial UKA. Goal of this study therefore was to assess outcomes of robotic-assisted medial UKA in a large cohort of patients at short-term follow-up. This multicenter study with IRB approval examines the survivorship and satisfaction of this robotic-assisted procedure coupled with an anatomically designed UKA implant at a minimum of two-year follow-up. A total of 1007 patients (1135 knees) underwent robotic-assisted surgery for a medial UKA from six surgeons at separate institutions in the United States. All patients received a fixed-bearing metal backed onlay implant as the tibial component between March 2009 and December 2011 (Figure 1). Each patient was contacted at minimum two-year follow-up and asked a series of five questions to determine implant survivorship and patient satisfaction. Survivorship analysis was performed using Kaplan-Meier method and worst-case scenario analysis was performed whereby all patients were considered as revision when they declined study participation. Revision rates were compared in younger and older patients (age cut-off 60 years) and in patients with different body mass index (body mass index cut-off 35 kg/m2). Two-sided chi-square tests were used to compare these groups.INTRODUCTION
METHODS
There is a paucity of studies analyzing the rates of revision total knee arthroplasty in diabetic patients stratified by glycated hemoglobin levels. The purpose of this study was to: 1) determine the incidence of revision TKA; 2) correlate the percent of glycated hemoglobin with incidence of revision; and 3) determine the cause of revision in diabetic patients stratified by glycated hemoglobin level. We utilized a national private payer dataset within the PearlDiver database from 2007 to 2015 quarter 1 to determine who had diabetes and underwent TKA. There were 424,107 patients who were included in the analysis. We determined the incidence of revision TKA in the overall cohort, in addition to stratifying the incidence by glycated hemoglobin levels. To determine the effect of glycated hemoglobin levels on revision TKA rate, we performed a correlation analysis between the level of glycated hemoglobin and the incidence of revision TKA. We performed descriptive statistics of the underlying cause of revision TKA in both the overall and stratified cohortsIntroduction
Methods
It is well established that diabetic patients undergoing total knee arthroplasty (TKA) are more susceptible to infection, problematic wound healing, and have overall higher complication rates. However there is a paucity in current literature investigating the effects of hypoglycemia on TKA. The purpose of this study was to determine the effect of hypoglycemia on TKA revision (rTKA) incidence analyzing a national private payer database for procedures performed between 2007 and 2015 Q1. A retrospective review of a national private payer database within the PearlDiver Supercomputer application (Warsaw, IN, USA) for patients undergoing TKA with blood glucose levels ranging from 20 to 219 mg/ml, in increments of 10 mg/ml, was conducted. Patients who underwent TKA were identified by Current Procedural Terminology (CPT)-27447 and International Classification of Disease (ICD) code 81.54. Glucose ranges were identified by filtering for lab identifier values unique to the PearlDiver database. Revision TKA and causes for revision, including mechanical loosening, failure/break, periprosthetic fracture, osteolysis, infection, pain, arthrofibrosis, instability, and trauma) were identified with CPT and ICD-9 Codes. Statistical analysis of this study was primarily descriptive.Introduction
Methods
Total knee arthroplasty (TKA) is currently one of the most common elective surgical procedures in the United States. The increase in the proportion of younger patients in receipt of surgery, in concert with a dramatic rise in the incidence of obesity, has contributed to the on-going, exponential increase in the number of arthroplasties performed annually. Despite materials advances for implants, the U.S. revision burden has remained static for the last decade. According to the 2013 CMS MEDPAR file the typical CMS reimbursement falls far short of costs incurred by the hospital, resulting in an average net loss of revenue of $9,539; and over 90% of hospitals lose money for every revision case performed. Today, approximately 5% of all primaries performed will result in an early revision (< 3 years). In order to understand ways with which to mitigate the incidence of early revision due to mechanical complications, a multicentric group of sensor-assisted patients was follow-up out to 3 years. In this study, 278 sensor-assisted patients were followed out to 3 years. The intraoperative devices used in this study contain microsensors and a processing unit. Kinetic and center of load location data are projected, in real-time, to a screen. Because of the wireless nature of the intraoperative sensors, the patella can be reduced, and kinematic data can be evaluated through the range of motion. For each patient, the soft-tissue envelope was balanced to within a mediolateral differential of 15 lbf., through the ROM, as per the suggestion of previously reported literature. The average patient profile indicates: age = 69.7 years, BMI = 30.4, gender distribution = 36% male/64% female. Any adverse event within the 3-year follow-up interval was captured. By 3 years, 1 patient in this population has required revision surgeon due to mechanical complicatons. Overall adverse events included: pain in hip (3), pain in contralateral knee (2), wound drainage (3), DVT (1), death (1), stiffness in operative knee (2), infection (3), global pain (2), back pain (2). Based on the average reported number of early revisions that occur in the U.S. (5% of primaries), it was anticipated for this patient group to require approximately 13 revisions by the 3-year follow-up interval. Using 2013 CMS MEDPAR data, these 13 revisions would have resulted in $124,007 cost-to-hospital. However, only 1 revision (0.4%) was observered, therefore $114,468 in additional costs were spared for the aggregate of participating hospitals. This data suggests that the incorporation of kinetic sensors in TKA may assist the surgeon in achieving soft-tissue balance and thereby avoiding adverse mechanical complications that require surgical intervention.
For nearly 58% of total knee arthroplasty (TKA) revisions, the reason for revision is exacerbated by component malalignment. Proper TKA component alignment is critical to functional outcomes/device longevity. Several methods exist for orthopedic surgeons to validate their cuts, however, each has its limitations. This study developed/validated an accurate, low-cost, easy to implement first-principles method for calculating 2D (sagittal/frontal plane) tibial tray orientation using a triaxial gyroscope rigidly affixed to the tibial plateau of a simulated leg jig and validated 2D tibial tray orientation in a human cadaveric model. An initial simulation assessed error in the sagittal/frontal planes associated with all geometric assumptions over a range of positions (±10°, ±10°, and −3°/0°/+3° in the sagittal, frontal, and transverse planes, respectively). Benchtop experiments (total positions - TP, clinically relevant repeated measures - RM, novice user - NU) were completed using a triaxial gyroscope rigidly affixed to and aligned with the tibial tray of the fully adjustable leg-simulation jig. Finally, two human cadaveric experiments were completed. A similar triaxial gyroscope was mounted to the tibial tray of a fresh frozen human cadaver to validate sagittal and frontal plane tibial tray orientation. In cadaveric experiment one, three unique frontal plane shims were utilized to measure changes in frontal plane angle. In cadaveric experiment two, measurements using the proprosed gyroscopic method were compared with computer navigation at a series of positions. For all experiments, one rotation of the leg was completed and gyroscopic data was processed through a custom analysis algorithm.Introduction
Methods
Total knee arthroplasty (TKA) patients are consistently reported to be less satisfied than total hip arthroplasty (THA) patients. A patient's perception of success of his/her own total knee is dictated by their levels of post-operative pain and function, and many return to follow-up visits with inexplicable pain and stiffness that contradict favorable radiographic results. Several of these chief complaints that contribute to dissatisfaction are associated with soft-tissue imbalance. Therefore, in an effort to thoroughly understand the post-operative impact of soft-tissue balance on satisfaction, a multicenter study was conducted to evaluate the satisfaction outcomes of quantifiably balanced patients. In this study, 102 sensor-assisted patients were followed out to 3 years. The intraoperative devices used in this study project kinetic loading (lbf.) and center of load location data, in real-time, to a screen. Because of the wireless nature of the intraoperative sensors, the patella can be reduced, and kinematic data can be evaluated through the range of motion. The target balance window that was used in this study has been previously reported in literature and includes: 1) a mediolateral differential of 15 lbf., through the ROM, and 2) Sagittal plane stability as determined by a posterior drawer analysis. A robust, face-validated satisfaction survey was administered at 3-year follow-up and included 7 questions with answers on a 5-point Likert scale. At 3 years, post-operatively, 97.2% of this patient group reported being “satisfied” to “very satisfied” with their procedure, in comparison to the 81% average TKA satisfaction reported in literature (df = 11). The comparative literature included annual satisfaction intervals from 1 to 5 years (n = 33,775) which is comparable to the interval reported in this patient group. The sensor-assisted patient group exhibited a 16% increase in the proportion of satisfaction over what is currently reported in the comparative literature (p = 0.001). Despite the success rate of TKA, unfavorable patient-reported satisfaction continues to present a problem for operative recipients and surgeons. Because satisfaction is dependent upon several variables – including pain, function, and activity levels – the satisfaction survey used in this study represents a more accurate account of patient perception than many traditional surveys. It was shown that sensor-balanced TKA patients exhibited a 16% increase in the proportion of those reporting being “satisfied” to “very satisfied”, over the average satisfaction reported in literature. Allowing the surgeon to quantitatively balance the soft-tissue envelope, dynamically, has continued to a significant decrease in the proportion of dissatisfaction.
The rate of technological innovation in procedural total knee arthroplasty has left little time for critical evaluation of a new technology before the adoption of even newer modalities. With more drastic financial restrictions being placed on operating room spending, orthopaedic surgeons are now required to provide excellent results on a budget. It is integral that both clinical efficacy and cost-effectiveness of these intraoperative technologies be fully understood in order to provide patients with effectual, economically conscious care. The purpose of this qualitative analysis of literature was to evaluate clinical and economic efficacy of the three most prominent technologies currently used in TKA: computer navigation, patient-specific instrumentation, and kinetic sensors. Three hundred and ninety one publications were collected; 100 were included in final qualitative analysis. Criteria for inclusion in the analysis was defined only insofar as that each piece assessed one of the above listed aspects of the three technologies Literature included in the final evaluation contained background information on each respective technology, clinical outcomes, revision rates, and/or cost analyses. All comparisons were conducted in a strictly qualitative manner, and no attempts were made to conduct interstudy statistical analyses due to the high level of variability in methodology and data collected.Introduction
Methods
Instability after total knee arthroplasty (TKA) represents, in excess of, 7% of reasons for implant failure. This mode of failure is correlated with soft-tissue imbalance, and has continued to be problematic despite advances in implant technology. Thus, understanding the options available to execute safe and effective soft-tissue release is critical to mitigating future complications due to instability. This study aimed to use intraoperative sensors to evaluate a multiple needle puncturing technique (MNPT), in comparison with traditional transection-based release, to determine its biomechanical and clinical efficacy. Seventy-five consecutive, cruciate-retaining TKAs were performed, as part of an 8-site multicenter study. All procedures were performed with the use of an intraoperative sensor to ensure quantitative balance, as per previously reported literature. Of the 75-patient cohort, 50 patients were balanced with the MNPT; 20 patients were balanced with traditional transection. All patients were followed out to 1-year, and administered KSS, WOMAC, and satisfaction. Alignment and ROM was captured for all patients, pre-operatively and at the 1-year follow-up interval.Introduction
Methods
Recent advances in 3D printing enable the use of custom patient-specific instruments to place drill guides and cutting slots for knee replacement surgery. However, such techniques limit the ability to intra-operatively adjust an implant plan based on soft-tissue tension and/or joint pathology observed in the operating room, e.g. cruciate ligament integrity. It is hypothesized that given the opportunity, a skilled surgeon will make intra-operative adjustments based on intra-operative information not captured by the hard tissue anatomy reconstructed from a pre-operative CT scan or standing x-ray. For example, tibiofemoral implant gaps measured intra-operatively are an indication of soft-tissue tension in the patient's knee, and may influence a surgeon to adjust implant position, orientation or size. This study investigates the frequency and magnitude of intra-operative adjustments from a single orthopedic surgeon during 38 unicondylar knee arthroplasty (UKA) cases. For each patient, a pre-operative plan was created based on the bony anatomy reconstructed from the pre-operative CT. This plan is analogous to a plan created with patient-specific cutting blocks or customized implants. With robotic technology that utilizes pre-operative imaging, intra-operative navigation and robotic execution, this “anatomic” plan can be fine-tuned and adjusted based on the soft tissue envelop measured intra-operatively. The relative positions of the femur and the tibia are measured intra-operatively under a valgus load (for medial UKA, varus load for lateral UKA) for each patient from extension to deep knee flexion and used to compute the predicted space between the implants (gaps) throughout flexion. The planned position, orientation and size of the components can then be adjusted to achieve an optimal dynamic ligament balance prior to any bony cuts. This is the plan that is then executed under robotic guidance. Intra-operative adjustments are defined as any size, position or orientation changes occurring intra-operatively to the pre-operative anatomic plan.Introduction
Methods
Aseptic loosening has been reported to be the most common, contemporary mode of total knee arthroplasty failure. It has been suggested that the etiology of revision due to loosening can be attributed, in part, to joint imbalance and the variability inherent in standard surgical techniques. Due to the high prevalence of revision, the purpose of this study was to quantify the change in kinetic loading of the knee joint before versus after the application of the final cement-component complex. Ninety-two consecutive, cruciate-retaining TKAs were performed, between March 2014 and June 2014, by two collaborating surgeons. Two different knee systems were used, each with a different viscosity cement type (either medium viscosity or high viscosity). All knees were initially balanced using a microelectronic tibial insert, which provides real-time feedback of femoral contact points and joint kinetics. After the post-balance loads were captured, and the surgeon was satisfied with joint balance, the final components were cemented into place, and the sensor was re-inserted to capture any change in loading due to cementing technique.Introduction
Methods
High BMI has been classically regarded as a contraindication for unicompartmental knee arthroplasty (UKA) as it can potentially lead to poor clinical outcomes and a higher risk of failure. In recent years, UKA has increased in popularity and, as a result, patient selection criteria are beginning to broaden. However, UKA performed manually continues to be technically challenging and surgical technique errors may result in suboptimal implant positioning. UKA performed with robotic assistance has been shown to improve component positioning, overall limb alignment, and ligament balancing, resulting in overall improved clinical outcomes. The purpose of this study is to examine the effect of high BMI in patients receiving UKA with robotic assistance. 1007 patients (1135 knees) were identified in an initial and consecutive multi-surgeon multi-center series receiving robotically assisted medial UKA, with a fixed bearing metal backed onlay tibial component. As part of an IRB approved study, every patient in the series was contacted at a minimum two year (±2 months) follow up and asked a series of questions to determine implant survivorship and satisfaction. 160 patients were lost to follow up, 35 patients declined to participate, and 15 patients were deceased. 797 patients (909 knees) at a minimum two year follow up enrolled in the study for an enrollment rate of 80%. 45% of the patients were female. The average age at time of surgery was 69.0 ± 9.5 (range: 39–93). BMI data was available for 887 knees; the average BMI at time of surgery was 29.4 ± 4.9. Patients were stratified in to five categories based on their BMI: normal (< 25; 16%), overweight (25–30; 46%), obese class I (30–35; 25%), obese class II (35–40; 11%) and obese class III (>40; 2%).Introduction
Methods
Achieving balance in TKA is critical in assuring favorable outcomes. But, in order to achieve quantifiably balanced loading values, is it more advantageous to make bony corrections or release soft-tissue? The answer to this question will be paramount in evaluating the most appropriate surgical techniques for use with new dynamic technology, thereby maximizing favorable clinical outcomes. Therefore, the purpose of this investigation was to evaluate a possible quantitative loading threshold, using intraoperative sensors, which may dictate surgical correction of bone versus soft-tissue release. A retrospective analysis of 122 multicenter patients, in receipt of sensor-assisted primary TKA, was conducted. 40 lbs. was used as a threshold, above which bone was corrected; below which soft-tissue was corrected. All patients were categorized in to the following groups: Group A – candidates for bony correction, but received soft-tissue correction; Group B – candidates for soft-tissue/receiving soft-tissue; Group C – candidates for bony correction/receiving bony correction.INTRODUCTION
METHODS
Patient-reported satisfaction is a critical measure in understanding the clinical success of total knee arthroplasty. Yet, satisfaction levels in TKA patients are generally lower than THA patients; and surgeon-patient agreeability regarding clinical success is typically in discordance. Thus, the purpose of this evaluation was to report on the one-year satisfaction data of a group of sensor-assisted TKA patients, and compare that data to the average satisfaction reported in literature, as measured by a meta-analysis. One hundred and thirty five patients received TKA utilizing intra-operative sensing technology to evaluate soft-tissue balance as part of a prospective multicenter study. Patients were classified by two groups: “balanced” and “unbalanced”. Quantitative “balance” was defined as a mediolateral intercompartmental loading difference of ≤ 15 pounds; all loading exceeding 15 pounds was classified as “unbalanced”. At the one-year follow-up visit, a 7-question patient satisfaction survey was administered. The answering schema of this survey was modeled using a modified five-point Likert scale, ranging from “True” to “False” (or “Very Satisfied” to “Very Dissatisfied,” where appropriate). A meta-analysis of literature was performed and studies selected for inclusion in this analysis were required to meet the following criteria: all patients were in receipt of a primary TKA; satisfaction data was collected post-operatively; and the proportion of patients who were “satisfied” to “very satisfied” was statistically described.INTRODUCTION
METHODS
The cost associated with the TKA revision burden is projected to reach 13 billion dollars, annually. Complications reported by post-TKA patients include: pain (44%, multilocational), sensation of instability (21% reason for revision), and joint stiffness (17% reason for revision); problems that may be attributed to soft-tissue imbalance. One of the possible reasons for the substantial prevalence of such complications is the subjectivity associated with defining soft-tissue balance. A priority must be placed on developing new objective methods with which to avoid costly post-operative complications, including the integration of intraoperative sensing technology. The purpose of this evaluation was to report on the disparity between the patient-reported outcomes scores of quantitatively balanced versus unbalanced patients, at 1-year, using a group of 135 multicenter patients. 135 prospective patients, from 8 U.S. sites, have had primary TKA performed with the use of intraoperative sensors. Patients were classified by two groups: “balanced” and “unbalanced”. Quantitative “balance” was defined as a mediolateral intercompartmental loading difference of ≤ 15 pounds; all loading exceeding 15 pounds was classified as “unbalanced”. For all patients, the following kinematic data was captured: varus/valgus stability, anteroposterior stability, flexion contracture (if any), extension lag (if any), anatomic alignment, and ROM. Also at each clinical follow-up visit, activity levels and two patient-reported outcomes measures were administered, including: the American Knee Society Score (KSS), and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC).INTRODUCTION
METHODS
Total knees today are performed with the use of standard trials that the surgeon utilises to define appropriate implant rotation, range of motion, and soft tissue balance. This “feel” based approach is very subjective, and lacks a quantifiable approach to interpret our intra-operative knee assessment. Sensor-based trials are embedded into the specific knee designed tibial trial, and wirelessly displays data related to the implant's position and ligament tension. The surgeon can now identify malrotation, soft tissue imbalance, and instability through a full ROM. The surgeon can see dynamic responses to ligament releases, alignment changes, and implant adjustments. As Insall taught us; a TKR is a soft tissue procedure, and a “balanced” knee will demonstrate improved outcomes and greater patient satisfaction. Smart Trials allow us to discuss how our intra-operative techniques affect our patient's outcomes. This surgery will utilise Smart Trials during a Cruciate Retaining TKR.
