Random forest identifies predictors of discharge destination following total shoulder arthroplasty.

Background: Machine learning algorithms are finding increasing use in prediction of surgical outcomes in orthopedics. Random forest is one of such algorithms popular for its relative ease of application and high predictability. In the process of sample classification, algorithms also generate a list of variables most crucial in the sorting process. Total shoulder arthroplasty (TSA) is a common orthopedic procedure after which most patients are discharged home. The authors hypothesized that random forest algorithm would be able to determine most important variables in prediction of nonhome discharge. Methods: Authors filtered the National Surgical Quality iImprovement Program database for patients undergoing elective TSA (Current Procedural Terminology 23472) between 2008 and 2018. Applied exclusion criteria included avascular necrosis, trauma, rheumatoid arthritis, and other inflammatory arthropathies to only include surgeries performed for primary osteoarthritis. Using Python and the scikit-learn package, various machine learning algorithms including random forest were trained based on the sample patients to predict patients who had nonhome discharge (to facility, nursing home, etc.). List of applied variables were then organized in order of feature importance. The algorithms were evaluated based on area under the curve of the receiver operating characteristic, accuracy, recall, and the F-1 score. Results: Application of inclusion and exclusion criteria yielded 18,883 patients undergoing elective TSA, of whom 1813 patients had nonhome discharge. Random forest outperformed other machine learning algorithms and logistic regression based on American Society of Anesthesiologists (ASA) classification. Random forest ranked age, sex, ASA classification, and functional status as the most important variables with feature importance of 0.340, 0.130, 0.126, and 0.120, respectively. Average age of patients going to facility was 76 years, while average age of patients going home was 68 years. 78.1% of patients going to facility were women, while 52.7% of patients going home were. Among patients with nonhome discharge, 80.3% had ASA scores of 3 or 4, while patients going home had 54% of patients with ASA scores 3 or 4. 10.5% of patients going to facility were considered of partially/totally dependent functional status, whereas 1.3% of patients going home were considered partially or totally dependent (P value < .05 for all). Conclusion: Of various algorithms, random forest best predicted discharge destination following TSA. When using random forest to predict nonhome discharge after TSA, age, gender, ASA scores, and functional status were the most important variables. Two patient groups (home discharge, nonhome discharge) were significantly different when it came to age, gender distribution, ASA scores, and functional status.

Serum CD64 as a Marker for Chronic Periprosthetic Joint Infection.

Background: Serum cluster of differentiation 64 (CD64) has emerged as a diagnostic test for musculoskeletal infections. The purpose of this study was to evaluate the utility of serum CD64 in diagnosing periprosthetic joint infections (PJIs) compared to conventional markers like white blood count (WBC), C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and interleukin-6 (IL-6). Methods: A prospective case-control study on patients undergoing revision hip or knee arthroplasty surgery >6 weeks after their index surgery was performed at a single institution. Whole blood samples were drawn within 24 hours prior to revision surgery for white blood count, ESR, CRP, IL-6, and CD64. Intraoperative cultures were obtained during the revision, and PJI was defined using the major criteria from the 2018 Musculoskeletal Infection Society criteria. Two-sample Wilcoxon rank-sum test and Fisher's exact test were used to determine if there were significant differences in serum laboratory values between patients with and without infection. The sensitivity, specificity, positive predictive value (PPV), negative predictive value, and accuracy of each test were calculated. Results: With an average age of 67 years, 39 patients with 15 revision THAs and 24 TKAs, were included. 19 patients (48.7%) were determined to have PJI. Patients with PJI had significantly higher CD64 (P = .036), CRP (P = .016), and ESR (P = .045). CD64 had the highest specificity (100%) and PPV (100%), moderate accuracy (69.2%), but low sensitivity (37.0%) and negative predictive value (62.5%). Conclusions: Given the high specificity, PPV, and accuracy, CD64 may be an excellent confirmatory test to help diagnose PJI.

Impact of Antibiotic-Impregnated Spacer Design in the Treatment of Periprosthetic Hip Infection.

Background: Although 2-stage exchange arthroplasty, consisting of temporary insertion of an antibiotic-impregnated cement spacer (AICS), is considered the standard of care for chronic periprosthetic joint infection (PJI) in total hip arthroplasty (THA), a consensus on the AICS design has not yet been established. Ceramic-on-polyethylene AICSs (Poly-AICS) are theorized to cause less pain and better function than cement-on-bone AICS (CemB-AICS) but use non-antibiotic-impregnated components that may harbor bacteria. This study evaluates the impact of spacer design on infection-free survivorship following THA reimplantation as well as pain and function during the interim AICS stage. Methods: A retrospective review was performed of all cases of THA PJI treated with either Poly-AICS or CemB-AICS at a single high-volume academic center. Data were collected until the final follow-up after THA reimplantation with an average follow-up duration of 2.6 years. The primary outcome was infection-free survivorship after the final reimplantation. Secondary outcomes included postoperative pain scores, opioid use, time to ambulation, length of stay, complications, and discharge disposition. Results: A total of 99 cases (67 CemB-AICS; 32 Poly-AICS) were included. There were no baseline differences between the 2 groups. There were no differences in infection-free survivorship after reimplantation in survivorship curve comparisons (P = .122) and no differences in postoperative inpatient pain scores, opioid use, length of stay, time to ambulation, complications, or discharge disposition during the AICS stage. Conclusions: Patients with THA PJI treated with Poly-AICS did not have worse infection-related outcomes despite the use of non-antibiotic-impregnated components but also did not appear to have less pain or improved function during the early AICS stage.

Comparable Torque to Failure Using the Simple Stich Versus the Figure-of-Eight Configuration for Hip Capsular Closure Following an Interportal Capsulotomy: A Cadaveric Study.

PURPOSE: To measure and compare the torque to failure and stiffness of the capsular repair construct consisting of four-suture simple stitches to a two-figure of eight stitches repair construct in external rotation following an interportal capsulotomy. METHODS: Six pairs of fresh-frozen cadaveric hemipelves were divided into two capsular repair groups. All hips underwent a 40-mm interportal capsulotomy from the 12 o'clock position to the 3 o'clock position. Capsular closure was performed using either the two stitches in a figure of eight or with four simple stitches. Afterward, each hemipelvis was securely fixed to the frame of a mechanical testing system with the hip in 10° of extension and externally rotated to failure. Significance was set at P < .05. RESULTS: The average failure torque was 86.2 ± 18.9 N·m and 81.5 ± 8.9 N·m (P = .57) for the two stitches in a figure of eight and the four simple stitches, respectively. Failure stiffness was also not statistically different between groups and both capsular closure techniques failed at similar degrees of rotation (P = .65). CONCLUSION: Hip capsular repair using either the four simple stitch or two-figure of eight configurations following interportal capsulotomy demonstrated comparable failure torques and similar stiffness in a cadaveric model. CLINICAL RELEVANCE: Adequate and comprehensive capsular management in hip arthroscopy is critical. Capsular repair following capsulotomy in femoroacetabular impingement surgery has been associated with higher patient-reported outcomes when compared to capsulotomy without repair. Therefore, determining which capsular closure construct provides the higher failure torque is important.

Calculating Intraoperative Fluid Deficit to Prevent Abdominal Compartment Syndrome in Hip Arthroscopy.

Abdominal compartment syndrome (ACS) is a rare but potentially fatal complication that can occur during hip arthroscopy. This usually occurs as a result of arthroscopic fluid passing into the retroperitoneal space through the psoas tunnel. From the retroperitoneal space, the fluid can then enter the intraperitoneal space through defects in the peritoneum. Previous studies have identified female sex, iliopsoas tenotomy, pump pressure, and operative time as potential risk factors for fluid extravasation. We present a method to measure intraoperative fluid deficit during hip arthroscopy to alert surgeons to possible ACS. Our proposed technique requires diligent intraoperative monitoring of fluid output through various suction devices, including suction canisters, puddle vacuums, and suction mats. The difference is then calculated from the fluid intake from the arthroscopic fluid bags. If the difference is greater than 1500 mL, then the anesthesiologist and circulating nurse are instructed to examine the abdomen for distension every 15 minutes. This, combined with other common symptoms such as hypotension and hypothermia, should alert the surgical team to the development of ACS. Despite limitations to this technique, this approach offers an objective method to calculate intra-abdominal fluid extravasation.

Effect of Patient Height and Sex on the Patellar Tendon and Anterior Cruciate Ligament.

BACKGROUND: Graft-tunnel mismatch is an avoidable complication in anterior cruciate ligament (ACL) reconstruction. Patient height and sex may be predictors of patellar tendon length (PTL) and intra-articular ACL length (IAL). Understanding these relationships may assist in reducing graft-tunnel mismatch during ACL reconstruction with bone-patellar tendon-bone (BTB) autograft. PURPOSE: To determine the association of patient height and sex with PTL and IAL. STUDY DESIGN: Cross-sectional study; Level of evidence, 3. METHODS: Magnetic resonance imaging (MRI) studies were obtained on the healthy knees of 100 male and 100 female patients. Patients with prior surgery, open physes, significant degenerative changes, ACL rupture, or extensor mechanism injury were excluded. Three independent readers measured PTL, IAL, and Caton-Deschamps Index (CDI) on MRI. Bivariate and linear regression analysis was performed to detect the association of anthropometric data with anatomic parameters measured on MRI studies. RESULTS: The mean age and body mass index were not significantly different between the male and female patients; however, male patients were significantly taller than female patients (1.75 vs 1.72 m, respectively; P < .001). There was a substantial agreement between the 3 readers for all parameters (κ > 0.75). Overall, female patients had significantly longer PTL (47.38 vs 43.92 mm), higher CDI (1.146 vs 1.071), and shorter IAL (33.05 vs 34.39 mm) (P < .001 for all). Results of the linear regression analysis demonstrated that both height and female sex were predictive of longer PTL. Further, height was independently predictive of IAL but sex was not. CONCLUSION: PTL was correlated more with patient sex than height. IAL was also correlated with patient sex. Longer BTB grafts are expected to be harvested in female patients compared with male patients of the same height despite shorter IAL. These associations should be considered during BTB ACL reconstruction to minimize graft-tunnel mismatch.

Fascia Iliaca Regional Anesthesia in Hip Fracture Patients Revisited: Which Fractures and Surgical Procedures Benefit Most?

OBJECTIVES: Perioperative fascia iliaca regional anesthesia (FIRA) decreases pain in hip fracture patients. The purpose of this study is to determine which hip fracture types and surgical procedures benefit most. DESIGN: Prospective observational study compared with a retrospective historical control. PATIENTS/PARTICIPANTS: Patients older than 60 years who received perioperative FIRA were compared with a historical cohort not receiving FIRA. SETTING: This study was conducted at a Level 1 trauma center. MAIN OUTCOME MEASUREMENTS: The primary outcome was morphine milliequivalents (MME) consumed during the index hospitalization. Fracture pattern-specific preoperative and postoperative MME consumption and surgical procedure-specific postoperative MME consumption was compared between the FIRA and non-FIRA groups. RESULTS: A total of 949 patients were included in this study, with 194 (20.4%) patients in the prospective protocol group. There were no baseline differences between cohorts. Preoperatively, only femoral neck fracture patients receiving FIRA used fewer MME (P < 0.001). Postoperatively, femoral neck fracture patients receiving FIRA used fewer MME on postoperative day (POD) 1 (P = 0.027) and intertrochanteric fracture patients used fewer MME on POD1 and POD2 (P = 0.013; P = 0.002). Cephalomedullary nail patients receiving FIRA used fewer MME on POD1 and POD2 (P = 0.004; P = 0.003). Hip arthroplasty patients receiving FIRA used fewer MME on POD1 (P = 0.037). Percutaneous pinning and sliding hip screw patients had no significant MME reduction from FIRA. CONCLUSIONS: Preoperatively, patients with femoral neck fractures benefit most from FIRA. Postoperatively, both patients with femoral neck fractures and intertrochanteric fractures benefit from FIRA. Patients undergoing cephalomedullary nail fixation or hip arthroplasty benefit most from FIRA postoperatively. LEVEL OF EVIDENCE: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

A Comparison in Outcomes of Preoperative Single-shot versus Continuous Catheter Fascia Iliaca Regional Anesthesia in Geriatric Hip Fracture Patients.

BACKGROUND: Fascia iliaca nerve blocks relieve pain in geriatric hip fracture patients and can be administered via a single-shot or continuous catheter. We compared perioperative opioid consumption and pain scores between these two blocks. METHODS: We performed a prospective, observational cohort study, including geriatric hip fracture patients who received a preoperative block. We compared morphine milligram equivalent (MME) consumption and visual analog scale (VAS) pain scores between single-shot and continuous fascia iliaca blocks at multiple time points: preoperative and on postoperative (POD) day 0, 1, and 2. We compared the change in preoperative total and hourly opioid consumption before and after block placement within and between groups. Secondary outcomes included opioid related adverse events, length of stay, and readmission rates. RESULTS: 107 patients were analyzed, 66 received a single-shot and 41 a continuous block. No significant differences were found between both blocks at any time point for median MME consumption or pain scores. MME [IQR]: preoperative 20.5 [6.0,48.8] vs. 24.0 [8.8,48.0], p=0.95; POD0 6.0 [0.0,18.6] vs. 10.0 [0.0,14.0], p=0.52; POD1 12.0 [0.0,30.0] vs. 18.0 [5.0,24.0], p=0.69; POD2 6.0 [0.0,21.2] vs. 12.0 [0.0,24.0], p=0.54. VAS [IQR]: preoperative 4.0 [2.2,5.3] vs. 4.6 [3.2,5.3], p=0.34; POD0 1.3 [0.0,3.7] vs. 2.5 [0.0,3.6], p=0.73; POD1 2.9 [1.7,4.4] vs. 3.7 [1.5,4.7], p=0.59; POD2 2.4 [1.0,4.4] vs. 3.3 [1.9,4.2], p=0.18. Preoperative MME/hr significantly decreased after the block for both groups: 1.05 [0.0,2.2] to 0.0 [0.0,0.0], p < 0.001; 1.4 [0.6,3.1] to 0.0 [0.0,0.1], p < 0.001. The reduction in MME/hr between groups was not significantly different: 0.9 [0.0,1.9] vs. 1.4 [0.6,3.1], p = 0.067. We found no significant differences in secondary outcomes between groups. CONCLUSIONS: We report no differences in opioid use and pain scores between single-shot and continuous catheter fascia iliaca nerve blocks. Both blocks similarly reduce preoperative opioid consumption.

Time to Block: Early Regional Anesthesia Improves Pain Control in Geriatric Hip Fractures.

BACKGROUND: Fascia iliaca nerve blocks (FIBs) anesthetize the thigh and provide opioid-sparing analgesia for geriatric patients with hip fracture awaiting a surgical procedure. FIBs are recommended for preoperative pain management; yet, block administration is often delayed for hours after admission, and delays in pain management lead to worse outcomes. Our objective was to determine whether opioid consumption and pain following a hip fracture are affected by the time to block (TTB). We also examined length of stay and opioid-related adverse events. METHODS: This prospective cohort study included patients who were ≥60 years of age, presented with a hip fracture, and received a preoperative FIB from March 2017 to December 2017. Individualized care timelines, including the date and time of admission, block placement, and surgical procedure, were created to evaluate the effect that TTB and time to surgery (TTS) had on outcomes. Patterns among TTB, TTS, and morphine milligram equivalents (MME) were investigated using the Spearman rho correlation. For descriptive purposes, we divided patients into 2 groups based on the median TTB. Multivariable regression for preoperative MME and length of stay was performed to assess the effect of TTB. RESULTS: There were 107 patients, with a mean age of 83.3 years, who received a preoperative FIB. The median TTB was 8.5 hours. Seventy-two percent of preoperative MME consumption occurred before block placement (pre-block MME). A longer TTB was most strongly correlated with pre-block MME (rho = 0.54; p < 0.001), and TTS was not correlated. Patients with a faster TTB consumed fewer opioids preoperatively (12.0 compared with 33.1 MME; p = 0.015), had lower visual analog scale scores for pain on postoperative day 1 (2.8 compared with 3.5 points; p = 0.046), and were discharged earlier (4.0 compared with 5.5 days; p = 0.039). There were no differences in preoperative pain scores, postoperative opioid consumption, delirium, or opioid-related adverse events. Multivariate regression showed that every hour of delay in TTB was associated with a 2.8% increase in preoperative MME and a 1.0% increase in the length of stay. CONCLUSIONS: Faster TTB in geriatric patients with hip fracture may reduce opioid use, pain, and length of stay. LEVEL OF EVIDENCE: Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Biomechanical evaluation of PCL reconstruction with suture augmentation.

PURPOSE: The purpose of this study was to compare kinematics and patellofemoral contact pressures of all inside and transtibial single bundle PCL reconstructions and determine if suture augmentation further improves the biomechanics of either technique. METHODS: Cadaveric knees were tested with a posterior drawer force, and varus, valgus, internal and external moments at 30, 60, 90, and 120° of flexion. Displacement, rotation, and patellofemoral contact pressures were compared between: Intact, PCL-deficient, All-Inside PCL reconstruction with (AI-SA) and without (AI) suture augmentation, and transtibial PCL reconstruction with (TT-SA) and without (TT) suture augmentation. RESULTS: Sectioning the PCL increased posterior tibial translation (PTT) from intact at 60° to 120° of flexion, p < 0.001. AI PCL reconstruction improved stability from the deficient-state but had greater PTT than intact at 90° of flexion, p < 0.05. Adding suture augmentation to the AI reconstruction further reduced PTT to levels that were not statistically different from intact at all flexion angles. TT reconstructed knees had greater PTT than intact knees at 60, 90, and 120° of flexion, p < 0.01. Adding suture augmentation (TT-SA) improved posterior stability to PTT levels that were not statistically different from intact knees at 30, 60, and 120° of flexion. Patellofemoral pressures were highest in PCL-deficient knees at increased angles of flexion and were reduced after reconstruction, but this was not significant. CONCLUSION: In this time-zero study, both the all-inside and transtibial single bundle PCL reconstructions effectively reduce posterior translation from the deficient-PCL state. In addition, suture augmentation of both techniques provided further anterior-posterior stability.

Anterior cruciate ligament repair versus reconstruction: A kinematic analysis.

BACKGROUND: The purpose of this study was to compare the biomechanical properties of an anterior cruciate ligament (ACL) anatomic repair of a true femoral avulsion to an anatomic ACL reconstruction. It was hypothesized that the ACL repair and ACL reconstruction would have comparable biomechanical behavior when compared to the native knee. METHODS: Ten paired fresh-frozen cadaveric knees (n = 20) were used to investigate knee kinematics when an anterior drawer force, varus, valgus, internal, and external rotational moment were applied at 0, 15, 30, 45, 60, and 90 degrees of flexion. Displacement and rotation were recorded in the following conditions: ACL-intact, ACL-deficient, and ACL-repaired vs reconstructed. RESULTS: Sectioning of the ACL significantly increased anterior tibial translation (0°, 15°, 30° and 45°) compared to the intact state. The mean anterior displacement difference from intact was lower in the ACL-repaired knees compared to reconstructed knees at 30° and 90°. There were no significant differences between conditions in varus, valgus, internal, or external rotations. CONCLUSION: ACL repair and ACL reconstruction procedures restored knee anterior tibial translation in matched paired specimens. There were no differences in valgus, varus, internal, or external rotation. Although, ACL-repaired knees (avulsion model) demonstrated less anterior tibial translation when compared to ACL-reconstructed knees, this difference was less than one millimeter. Based on the findings of this study, repair and reconstruction procedures both restored anterior tibial translation in matched-pair specimens. This suggests that the initial functionality of both techniques is similar and that further clinical studies are needed to compare the long-term stability.

Increasing Use of Reverse Total Shoulder Arthroplasty for Proximal Humerus Fractures in Elderly Patients.

This study described surgical treatment patterns for proximal humerus fractures among elderly patients, focusing on reverse total shoulder arthroplasty (TSA), and evaluated how the type of fixation affects inpatient factors (cost, length of stay), transfusion rates, and patient disposition (home vs skilled nursing facility). With Nationwide Inpatient Sample data from 2011 to 2013, the authors identified patients 65 years and older who had proximal humerus fractures and divided them into 3 groups: (1) open reduction and internal fixation (ORIF); (2) hemiarthroplasty; and (3) reverse TSA. From 2011 to 2013, 38,729 surgically treated proximal humerus fractures were identified. The rate of reverse TSA increased 1.8-fold during this time, from 13% of operative cases in 2011 to 24% of operative cases in 2013 (P<.001). At the same time, the rates of hemiarthroplasty and ORIF decreased (hemiarthroplasty, from 28% to 21%; ORIF, from 59% to 55%). Although reverse TSA accounted for 32.2% of arthroplasty procedures for proximal humerus fractures in 2011, this value was 53.3% in 2013 (P<.001). In 2013, mean total hospital cost for reverse TSA was $24,154, which was significantly higher than that for ORIF ($16,269) or hemiarthroplasty ($19,175) (P<.001). In a multivariable model, patients undergoing reverse TSA were less likely than those undergoing hemiarthroplasty to be discharged to a skilled nursing facility (odds ratio, 0.75; P=.027). The national rate of reverse TSA nearly doubled from 2011 to 2013. As of 2013, reverse TSA replaced hemiarthroplasty as the most commonly performed arthroplasty procedure for proximal humerus fractures for patients 65 years and older. Patients undergoing reverse TSA were more likely than those undergoing hemiarthroplasty to be discharged home. [Orthopedics. 2017; 40(6):e982-e989.].