The Curtis Hand Injury Matrix Score: Determining the Need for Specialized Upper Extremity Care.

PURPOSE: Given the limited impact of transfer guidelines and the lack of comparative metrics for upper extremity trauma, we introduced the Curtis Hand Injury Matrix (CHIM) score to evaluate upper extremity injury acuity from the specialist perspective. Our goal was to evaluate the CHIM score as an indicator of complexity and specialist need by correlating the score with arrival mode, length of stay (LOS), discharge disposition, and procedure location. METHODS: We identified all hand and upper extremity emergency room visits at our institution in 2018 and 2019. On initial evaluation, our institution's hand surgery team assigned each patient an alphanumeric score with a number (1-5) and letter (A-H) corresponding to injury severity and pathology, respectively. Patients were divided into 5 groups (1-5) with lower scores indicating greater severity. We compared age, LOS, discharge disposition, procedure location, transfer status, and arrival mode between groups and assessed the relationships between matrix scores and discharge disposition, procedure performed, and LOS. RESULTS: There were 3,822 patients that accounted for 4,026 upper extremity evaluations. There were significant differences in LOS, discharge dispositions, procedure locations, transfer status, and arrival modes between groups. Patients with more severe scores had higher rates of admission and more operating room procedures. Higher percentages of patients who arrived via helicopter, ambulance, or transfer had more severe scores. Patients with more severe scores were significantly more likely to have a procedure, hospital admission, and longer hospital stay. CONCLUSIONS: The CHIM score provides a framework to catalog the care and resources required when covering specialized hand and upper extremity calls and accepting transfers. This clinical validation supports considering broader use. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic IV.

Understanding Surprise Out-of-Network Billing in Hand and Upper Extremity Care.

Surprise billing occurs when insured patients receive unexpected out-of-network charges and fees even when the emergency department, facility, or primary physician who provided care is in their insurance network. This issue is particularly relevant for hand surgery. The multidisciplinary nature of hand care and the number of ancillary services involved result in various levels at which out-of-network billing can be introduced, even when the hand surgeon is in-network for the patient. In addition, surprise billing is often associated with emergency department encounters, elective surgeries, and ambulance and helicopter transfers. In this article, we review surprise billing as it pertains to hand surgery. Little is known about surprise billing in hand care; however, we believe that these practices may substantially affect the patient population. We define key elements of surprise billing, review the literature, discuss the relevance and potential of surprise billing in hand surgery in various settings, and provide an overview of the status of health policy surrounding this practice. It is imperative for hand surgery as a field to understand the prevalence, operationalization, and policies of surprise billing better to prevent the exploitation of patients.

Surgical Debulking for Symptomatic Management of Calcinosis Cutis of the Hand and Upper Extremity in Systemic Sclerosis.

PURPOSE: To determine whether elective calcinosis debulking surgery of the hands and/or upper extremities is a safe and effective treatment for painful symptomatic scleroderma. Our hypothesis was that calcinosis debulking surgery would result in improvement in patient-reported pain and range of motion (ROM) with relatively little postoperative surgical pain for scleroderma patients. METHODS: We performed a retrospective review of scleroderma patients who underwent elective calcinosis debulking surgery by a single surgeon between August 2014 and August 2019. Patients were included if they had a documented diagnosis of limited or diffuse scleroderma and underwent elective or nonemergent hand or upper-extremity calcinosis debulking surgery with a minimum final follow-up of 12 months. Primary outcomes measured were preoperative to final follow-up changes in visual analog scale pain scores. Secondary outcomes were changes in numbness and ROM as well as in daily opioid requirements, postoperative opioids used to control surgical pain, and complications. RESULTS: Thirty-nine patients underwent calcinosis debulking surgeries on 41 upper extremities. Median final follow-up was 22 months (range, 13-60 months). Significant decreases occurred in visual analog pain scores (preoperative median, 5 [range, 0-10); final follow-up median, 0 [range, 0-8]) and improved patient-reported ROM in 15% (no change, 85%; worse, 0%). There was no significant preoperative to final follow-up difference in patient-reported numbness (improved, 5%; no change, 85%; and worse, 10%). Thirteen patients incurred 17 complications. CONCLUSIONS: Elective calcinosis debulking surgery of the hands and/or upper extremities in scleroderma decreased pain scores, improved patient-reported ROM in 15% of patients, and had no effect on patient-reported numbness at final follow-up. TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.

The Impact of Smoking on Early Postoperative Complications in Hand Surgery.

PURPOSE: Smoking is a prevalent modifiable risk factor that has been associated with adverse postoperative outcomes across numerous surgical specialties. We examined the impact of smoking on 30-day complications in patients undergoing hand surgery procedures. METHODS: The American College of Surgeons National Surgical Quality Improvement Program data sets were queried for patients who underwent common hand surgery procedures from 2011 to 2016. Cohorts were divided into smoking and nonsmoking and compared in terms of demographic characteristics, comorbidities, and postoperative complications. Multivariable logistic regression models were used to control for demographic characteristics and comorbidities in assessing the association between smoking and postoperative infections as well as other major and minor complications. RESULTS: We identified a cohort of 48,370 patients in the National Surgical Quality Improvement Program who underwent certain outpatient and inpatient hospital facility-based hand surgery procedures from 2011 to 2016. Up to 22% of these patients reported active smoking. Compared with nonsmokers, smokers were more likely to be younger and male and to have a lower body mass index. In addition, they were more likely to have a higher American Society of Anesthesiologists classification and to report dyspnea and chronic obstructive pulmonary disease. Multivariable logistic regression identified an independent association between smoking and major complications. Smoking was not significantly associated with minor complications. When regrouped by complication type, smoking was associated with infectious and wound healing complications. In subgroup analysis, smokers undergoing elective hand surgery had increased odds of wound healing complications but not major, minor, or infectious complications. CONCLUSIONS: Smokers may be at a significantly higher odds of certain complications compared with nonsmokers. For patients undergoing the elective procedures evaluated in this study, perioperative smoking may increase the risk of wound-healing complications. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic II.

The Management of Perioperative Immunosuppressant Medications for Rheumatoid Arthritis During Elective Hand Surgery.

PURPOSE: Rheumatoid arthritis (RA) is a destructive inflammatory disease that commonly involves joints of the hand and wrist. Different recommendations exist for continuing or discontinuing immunosuppressant medications during the perioperative time period. The purpose of our study was to determine whether continuing or discontinuing medications (steroids, nonbiological, and/or biological disease-modifying antirheumatic drugs [DMARDs]) were associated with an increased or decreased risk of postoperative complications. METHODS: We performed a single-center, retrospective review of a cohort of RA patients who had elective hand surgery by a single surgeon. Patients were included if they had a documented diagnosis of seropositive RA by a rheumatologist and had elective hand surgery and/or a disease-related surgical procedure involving the upper extremity between January 2008 and August 2018. We stratified patients into different groups for comparison by classes of immunosuppressant medications for managing RA. These classes included corticosteroids, nonbiological DMARDs, biological DMARDs, and/or no medications. Immunosuppressant medications were then compared with no medications for the incidence of postoperative overall complications. RESULTS: Eighty-eight patients had elective hand and/or upper extremity surgeries for RA. Mean patient age at the time of surgery (± SD) was 55 ± 13 years (range, 24-74 years). Of these 88 patients, 8 (9%) overall complications occurred. Complications were wound healing failures (n = 5), tendon rupture (n = 1), hematoma (n = 1), and surgical-site infection (n = 1). Perioperative medications included steroids (n = 31), nonbiological DMARDs (n = 68), biological DMARDs (n = 5), and no medication (n = 27). There were no significant differences in overall complications between patients on immunosuppressant medications and those on no medications. Median (interquartile range) follow-up was 11.5 months (5-25.8) (range, 2-74 months). CONCLUSIONS: We found that patients who continued or discontinued medications within 1 dosing interval of their usual dose perioperatively had similar rates of complications following elective hand surgery. TYPE OF STUDY/LEVEL OF EVIDENCE: Therapeutic IV.

Postoperative Free Flap Breast Protocol Optimizing Resources and Patient Safety.

BACKGROUND: As deep inferior epigastric artery perforator (DIEP) flaps have gained popularity in breast reconstruction, the postoperative care of these patients, including the appropriate hospital length-of-stay and the need for intensive care unit (ICU) admission, has become a topic of debate. At our institution, we have adopted a pathway that aims for discharge on postoperative day 3, utilizing continuous tissue oximetry without ICU admission. This study aims to evaluate outcomes with this pathway to assess its safety and feasibility in clinical practice. METHODS: A retrospective review was performed of patients undergoing DIEP flap breast reconstruction between January 2013 and August 2014. Data of interest included patient demographics and medical history as well as complication rates and date of hospital discharge. RESULTS: In total, 153 patients were identified undergoing 239 DIEP flaps. The mean age was 50 years (standard deviation [SD] = 10.2) and body mass index (BMI) 29.4 kg/m2 (SD = 5.2). Over the study period, the flap failure rate was 1.3% and reoperation rate 3.9%. Seventy-one percent of patients were discharged on postoperative day 3. Nine patients required hospitalization beyond 5 days. Theoretical cost savings from avoiding ICU admissions were $1,053 per patient. CONCLUSION: A pathway aiming for hospital discharge on postoperative day 3 without ICU admission following DIEP flap breast reconstruction can be feasibly implemented with an acceptable reoperation and flap failure rate.

The Effects of Medicaid Expansion on Triage and Regional Transfer After Upper-Extremity Trauma.

PURPOSE: Underinsured hand trauma patients are more likely to be transferred to quaternary care centers, which burdens these patients and centers. By increasing insurance coverage, care for less severe upper-extremity injuries may be available closer to patients' homes. We evaluated whether the 2014 expansion of Medicaid in Maryland under the Affordable Care Act decreased the number of uninsured upper-extremity trauma patients and the volume of unnecessary emergency trauma visits at our hand center. METHODS: We identified all upper-extremity trauma patients between 2010 and 2017 at our hand trauma referral center. Injury severity was classified based on the need for subspecialty care. Bivariate relations between insurance status and demographic covariates, including injury type and distance, both before and after Medicaid expansion were evaluated. We used patient-level and multinomial logistic regression models to evaluate changes in payer and transfer appropriateness. RESULTS: We studied 12,009 acute upper-extremity trauma patients. With Medicaid expansion, the percentage of trauma patients with Medicaid coverage increased from 15% to 24%, with a decrease in uninsured from 31% to 24%. After Medicaid expansion, non-transfer patient appropriateness decreased and appropriateness of transfers remained consistent across all payers. The average distance patients traveled for care remained similar before and after expansion. CONCLUSIONS: Medicaid expansion significantly decreased the proportion of uninsured upper-extremity trauma patients. We identified no significant changes in the distances these patients traveled for specialized care. In addition, the appropriateness of transferred patients did not change significantly after expansion, whereas appropriateness of nontransferred patients actually declined after Medicaid expansion. CLINICAL RELEVANCE: This study indicates no notable change in adherence to transfer guidelines after expansion, and a possible increase in use of emergency services by newly insured patients.

Vascularized Bone Grafting for Reconstruction of Oncologic Defects in the Spine: A Systematic Review and Pooled Analysis of the Literature.

BACKGROUND: Resection of primary spinal tumors requires reconstruction for restoration of spinal column stability. Traditionally, some combination of bone grafting and instrumentation is implemented. However, delayed healing environments are associated with pseudoarthrodesis and failure. Implementation of vascularized bone grafting (VBG) to complement hardware may present a solution. We evaluated the use of VBG in oncologic spinal reconstruction via systematic review and pooled analysis of literature. METHODS: We searched PubMed/MEDLINE, Embase, Cochrane, and Scopus for studies published through September 2017 according to the PRISMA guidelines and performed a pooled analysis of studies with n > 5. Additionally, we performed retrospective review of patients at the Johns Hopkins Hospital that received spinal reconstruction with VBG. RESULTS: We identified 21 eligible studies and executed a pooled analysis of 12. Analysis indicated an 89% (95% confidence interval [CI]: 0.75-1.03) rate of successful union when VBG is employed after primary tumor resection. The overall complication rate was 42% (95% CI: 0.23-0.61) and reoperation rate was 27% (95% CI: 0.12-0.41) in the pooled cohort. Wound complication rate was 18% (95% CI: 0.11-0.26). Fifteen out of 209 patients (7.2%) had instrumentation failure and mean time-to-union was 6 months. Consensus in the literature and in the patients reviewed is that introduction of VBG into irradiated or infected tissue beds proves advantageous given decreased resorption, increased load bearing, and faster consolidation. Downsides to this technique included longer operations, donor-site morbidity, and difficulty in coordinating care. CONCLUSIONS: Our results demonstrate that complication rates using VBG are similar to those reported in studies using non-VBG for similar spinal reconstructions; however, fusion rates are better. Given rapid fusion and possible hardware independence, VBG may be useful in reconstructing defects in patients with longer life expectancies and/or with a history of chemoradiation and/or infection at the site of tumor resection.

Is Article Methodological Quality Associated With Conflicts of Interest?: An Analysis of the Plastic Surgery Literature.

PURPOSE: Conflicts of interest (COI) are an emerging area of discussion within the field of plastic surgery. Recently, several reports have found that research studies that disclose COI are associated with publication of positive outcomes. We hypothesize that this association is driven by higher-quality studies receiving industry funding. This study aimed to investigate the association between industry support and study methodological quality. METHODS: We reviewed all entries in Plastic and Reconstructive Surgery, Annals of Plastic Surgery, and Journal of Plastic, Reconstructive, and Aesthetic Surgery within a 1-year period encompassing 2013. All clinical research articles were analyzed. Studies were evaluated blindly for methodology quality based on a validated scoring system. An ordinal logistic regression model was used to examine the association between methodology score and COI. RESULTS: A total of 1474 articles were reviewed, of which 483 met our inclusion criteria. These articles underwent methodological quality scoring. Conflicts of interest were reported in 28 (5.8%) of these articles. After adjusting for article characteristics in the ordinal logistic regression analysis, there was no significant association between articles with COI and higher methodological scores (P = 0.7636). CONCLUSIONS: Plastic surgery studies that disclose COI are not associated with higher methodological quality when compared with studies that do not disclose COI. These findings suggest that although the presence of COI is associated with positive findings, the association is not shown to be driven by higher-quality studies.

Giant cell reparative granuloma of the pediatric cranium: case report and review of the literature.

PURPOSE: Giant cell reparative granulomas are rare bone tumors. Although benign, these tumors are locally destructive and can be highly vascular. They seldom occur in the cranial vault. We describe a multidisciplinary approach to a case of giant cell reparative granuloma of the cranium in a 3-year-old patient. CASE REPORT: A 3-year-old girl female referred to the pediatric neurosurgery department for evaluation of a retro-auricular mass. She had a history of recurrent otitis media with two subsequent courses of antibiotics without resolution. CT imaging revealed an expansive lesion located in the right mastoid region. Open surgical biopsy revealed a hemorrhagic tumor consistent with a giant cell reparative granuloma. Angiography identified a hypervascular tumor blush that was supplied by the occipital artery. Preoperative transcatheter embolization was performed followed by a multidisciplinary surgical resection and reconstruction. Blood loss was minimal, and the patient recovered well after surgery. CONCLUSION: Preoperative endovascular embolization and a multidisciplinary intraoperative approach with primary resection and cranial vault reconstruction is an effective approach to hypervascular giant cell reparative granulomas.

Desmoplastic fibroma of the pediatric cranium: case report and review of the literature.

PURPOSE: Desmoplastic fibromas are primary bone tumors that seldom occur in the cranial bones. Furthermore, reports of desmoplastic fibromas of the skull in children are exceedingly rare. Although desmoplastic fibromas are histologically benign, they are locally aggressive and have a propensity to reoccur. Their radiographic appearance may mimic other more common central nervous system and bone neoplasms. There are only 19 reported cases of desmoplastic fibroma of the cranium in the literature, and only seven occurred in the pediatric age group. We present a case report of an 11-year-old female patient with a desmoplastic fibroma of the parieto-occipital region and review the literature. CASE REPORT: An 11-year-old female presented to the craniofacial clinic complaining of intermittent pain and a soft mass in the occipital region. There was a distant history of trauma to the region that did not require medical intervention. Computed tomography imaging revealed a lytic bone lesion overlying the sagittal sinus in the parieto-occipital region. Surgical resection with wide margins and immediate autologous reconstruction was performed. Pathological analysis revealed a desmoplastic fibroma. At 4 months of follow-up, no recurrence has been noted. CONCLUSION: Desmoplastic fibroma of the cranium is rare. Complete surgical resection with careful follow-up is the treatment of choice.

Readmission and Associated Factors in Surgical Versus Non-Surgical Management of Spinal Epidural Abscess: A Nationwide Readmissions Database Analysis.

STUDY DESIGN: Retrospective cohort study. OBJECTIVES: Spinal epidural abscess (SEA) is a rare but potentially life-threatening infection treated with antimicrobials and, in most cases, immediate surgical decompression. Previous studies comparing medical and surgical management of SEA are low powered and limited to a single institution. As such, the present study compares readmission in surgical and non-surgical management using a large national dataset. METHODS: We identified all hospital admissions for SEA using the Nationwide Readmissions Database (NRD), which is the largest collection of hospital admissions data. Patients were grouped into surgically and non-surgically managed cohorts using ICD-10 coding and compared using information retrieved from the NRD such as demographics, comorbidities, length of stay and cost of admission. RESULTS: We identified 350 surgically managed and 350 non-surgically managed patients. The 90-day readmission rates for surgical and non-surgical management were 26.0% and 35.1%, respectively (P < .05). Expectedly, surgical management was associated with a significantly higher charge and length of stay at index hospital admission. Surgically managed patients had a significantly lower risk of readmission for osteomyelitis (P < .05). Finally, in patients with a low comorbidity burden, we observed a significantly lower 90-day readmission rate for surgically managed patients (surgical: 23.0%, non-surgical: 33.8%, P < .05). CONCLUSION: In patients with a low comorbidity burden, we observed a significantly lower readmission rate for surgically managed patients than non-surgically managed patients. The results of this study suggest a lower readmission rate as an advantage to surgical management of SEA and emphasize the importance of SEA as a not-to-miss diagnosis.

Development of a machine learning algorithm to identify total and reverse shoulder arthroplasty implants from X-ray images.

Introduction: Demand for total shoulder arthroplasty (TSA) has risen significantly and is projected to continue growing. From 2012 to 2017, the incidence of reverse total shoulder arthroplasty (rTSA) rose from 7.3 cases per 100,000 to 19.3 per 100,000. Anatomical TSA saw a growth from 9.5 cases per 100,000 to 12.5 per 100,000. Failure to identify implants in a timely manner can increase operative time, cost and risk of complications. Several machine learning models have been developed to perform medical image analysis. However, they have not been widely applied in shoulder surgery. The authors developed a machine learning model to identify shoulder implant manufacturers and type from anterior-posterior X-ray images. Methods: The model deployed was a convolutional neural network (CNN), which has been widely used in computer vision tasks. 696 radiographs were obtained from a single institution. 70% were used to train the model, while evaluation was done on 30%. Results: On the evaluation set, the model performed with an overall accuracy of 93.9% with positive predictive value, sensitivity and F-1 scores of 94% across 10 different implant types (4 reverse, 6 anatomical). Average identification time was 0.110 s per implant. Conclusion: This proof of concept study demonstrates that machine learning can assist with preoperative planning and improve cost-efficiency in shoulder surgery.

Can Natural Language Processing and Artificial Intelligence Automate The Generation of Billing Codes From Operative Note Dictations?

STUDY DESIGN: Retrospective Cohort Study. OBJECTIVES: Using natural language processing (NLP) in combination with machine learning on standard operative notes may allow for efficient billing, maximization of collections, and minimization of coder error. This study was conducted as a pilot study to determine if a machine learning algorithm can accurately identify billing Current Procedural Terminology (CPT) codes on patient operative notes. METHODS: This was a retrospective analysis of operative notes from patients who underwent elective spine surgery by a single senior surgeon from 9/2015 to 1/2020. Algorithm performance was measured by performing receiver operating characteristic (ROC) analysis, calculating the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC). A deep learning NLP algorithm and a Random Forest algorithm were both trained and tested on operative notes to predict CPT codes. CPT codes generated by the billing department were compared to those generated by our model. RESULTS: The random forest machine learning model had an AUC of .94 and an AUPRC of .85. The deep learning model had a final AUC of .72 and an AUPRC of .44. The random forest model had a weighted average, class-by-class accuracy of 87%. The LSTM deep learning model had a weighted average, class-by-class accuracy 0f 59%. CONCLUSIONS: Combining natural language processing with machine learning is a valid approach for automatic generation of CPT billing codes. The random forest machine learning model outperformed the LSTM deep learning model in this case. These models can be used by orthopedic or neurosurgery departments to allow for efficient billing.

Trends in the Charges and Utilization of Computer-Assisted Navigation in Cervical and Thoracolumbar Spinal Surgery.

STUDY DESIGN: Retrospective national database study. PURPOSE: This study is conducted to assess the trends in the charges and usage of computer-assisted navigation in cervical and thoracolumbar spinal surgery. OVERVIEW OF LITERATURE: This study is the first of its kind to use a nationwide dataset to analyze trends of computer-assisted navigation in spinal surgery over a recent time period in terms of use in the field as well as the cost of the technology. METHODS: Relevant data from the National Readmission Database in 2015-2018 were analyzed, and the computer-assisted procedures of cervical and thoracolumbar spinal surgery were identified using International Classification of Diseases 9th and 10th revision codes. Patient demographics, surgical data, readmissions, and total charges were examined. Comorbidity burden was calculated using the Charlson and Elixhauser comorbidity index. Complication rates were determined on the basis of diagnosis codes. RESULTS: A total of 48,116 cervical cases and 27,093 thoracolumbar cases were identified using computer-assisted navigation. No major differences in sex, age, or comorbidities over time were found. The utilization of computer-assisted navigation for cervical and thoracolumbar spinal fusion cases increased from 2015 to 2018 and normalized to their respective years' total cases (Pearson correlation coefficient=0.756, p =0.049; Pearson correlation coefficient=0.9895, p =0.010). Total charges for cervical and thoracolumbar cases increased over time (Pearson correlation coefficient=0.758, p =0.242; Pearson correlation coefficient=0.766, p =0.234). CONCLUSIONS: The use of computer-assisted navigation in spinal surgery increased significantly from 2015 to 2018. The average cost grossly increased from 2015 to 2018, and it was higher than the average cost of nonnavigated spinal surgery. With the increased utilization and standardization of computer-assisted navigation in spinal surgeries, the cost of care of more patients might potentially increase. As a result, further studies should be conducted to determine whether the use of computer-assisted navigation is efficient in terms of cost and improvement of care.

The Impact of Computer-Assisted Navigation on Charges and Readmission in Patients Undergoing Posterior Cervical Fusion Surgery.

STUDY DESIGN: Retrospective cohort study of 2016 Healthcare Cost and Utilization Project Nationwide Readmissions Database (NRD). OBJECTIVE: The aim was to evaluate cost and outcomes associated with navigation use on posterior cervical fusion (PCF) surgery patients. SUMMARY OF BACKGROUND DATA: Computer-assisted navigation systems demonstrate comparable outcomes with hardware placement and procedural speed compared with traditional techniques. Innovations in technology continue to improve surgeons' performance in complicated procedures, causing need to analyze the impact on patient care. METHODS: The 2016 NRD was queried for patients with PCF surgery ICD-10 codes. Cost and readmission rates were compared with and without navigation. Nonelective cases and patients below 18 years of age were excluded. Univariate analysis on demographics, surgical data, and total charges was performed. Lastly, multivariate analysis was performed to assess navigation's impact on cost and postoperative outcomes. RESULTS: A total of 11,834 patients were identified, with 137 (1.2%) patients undergoing surgery with navigation and 11,697 (98.8%) patients without. Average total charge was $131,939.47 and $141,270.1 for the non-navigation and navigation cohorts, respectively ( P =0.349). Thirty-day and 90-day readmission rates were not significantly lower in patients who received navigation versus those that did not ( P =0.087). This remained insignificant after adjusting for several variables, age above 65, sex, medicare status, mental health history, and comorbidities. The model adjusting for demographic and comorbidities maintained insignificant results of navigation being associated with decreased 30-day and 90-day readmissions ( P =0.079). CONCLUSIONS: Navigation use in PCF surgery was not associated with increased cost, and patients operated on with navigation did not significantly have increased routine discharge or decreased 90-day readmission. As a result, future studies must continue to evaluate the cost-benefit of navigation use for cervical fusion surgery. LEVEL OF EVIDENCE: Level III.

The Impact of Computer-Assisted Navigation on Charges and Readmission in Lumbar Spinal Fusion.

STUDY DESIGN: Retrospective National Database Study. OBJECTIVES: The purpose of this study is to evaluate the cost and patient outcomes associated with the utilization of computer-assisted navigation (CAN) utilization on patients undergoing lumbar spinal fusion. BACKGROUND: CAN systems have demonstrated comparable outcomes with instrumentation and procedural speed when compared with traditional techniques. In recent years, CAN systems have seen increased adoption in spinal surgery as they allow for better contextualization of anatomical structures with the goal of improving surgical accuracy and reproducibility. METHODS: The 2016 National Readmission Database was queried for patients with lumbar spinal fusion ICD-10 codes, with 2 subgroups created based on computer-aided navigation ICD-10 codes. Nonelective cases and patients below 18 years of age were excluded. Univariate analysis on demographics, surgical data, and total charges was performed. Postoperative complication rates were calculated based on diagnosis. Lastly, multivariate analysis was performed to assess navigation's impact on cost and postoperative outcomes. RESULTS: A total of 88,445 lumbar fusion surgery patients were identified. Of the total, 2478 (2.8%) patients underwent lumbar fusion with navigation utilization, while 85,967 (97.2%) patients underwent surgery without navigation. The average total charges were $150,947 ($150,058, $151,836) and $161,018 ($155,747, $166,289) for the non-CAN and CAN groups, respectively ( P <0.001). The 30-day readmission rates were 5.3% for the non-CAN cohort and 3.1% for the CAN cohort ( P <0.05). The 90-day readmission rates were 8.8% for the non-CAN cohort and 5.2% for the CAN cohort ( P <0.001). CONCLUSIONS: CAN use was found to be significantly associated with increased cost and decreased 30-day and 90-day readmissions. Although patients operated on with CAN had increased routine discharge and decreased readmission risk, future studies must continue to evaluate the cost-benefit of CAN. Limitations include ICD-10 codes for CAN utilization being specific to region of surgery, not to exact type. LEVEL OF EVIDENCE: Level III.

Identification of Anterior Cervical Spinal Instrumentation Using a Smartphone Application Powered by Machine Learning.

STUDY DESIGN: Cross-sectional study. OBJECTIVE: The purpose of this study is to develop and validate a machine learning algorithm for the automated identification of anterior cervical discectomy and fusion (ACDF) plates from smartphone images of anterior-posterior (AP) cervical spine radiographs. SUMMARY OF BACKGROUND DATA: Identification of existing instrumentation is a critical step in planning revision surgery for ACDF. Machine learning algorithms that are known to be adept at image classification may be applied to the problem of ACDF plate identification. METHODS: A total of 402 smartphone images containing 15 different types of ACDF plates were gathered. Two hundred seventy-five images (∼70%) were used to train and validate a convolution neural network (CNN) for classification of images from radiographs. One hundred twenty-seven (∼30%) images were held out to test algorithm performance. RESULTS: The algorithm performed with an overall accuracy of 94.4% and 85.8% for top-3 and top-1 accuracy, respectively. Overall positive predictive value, sensitivity, and f1-scores were 0.873, 0.858, and 0.855, respectively. CONCLUSION: This algorithm demonstrates strong performance in the classification of ACDF plates from smartphone images and will be deployed as an accessible smartphone application for further evaluation, improvement, and eventual widespread use.Level of Evidence: 3.

Development of a machine learning algorithm to predict intubation among hospitalized patients with COVID-19.

PURPOSE: The purpose of this study is to develop a machine learning algorithm to predict future intubation among patients diagnosed or suspected with COVID-19. MATERIALS AND METHODS: This is a retrospective cohort study of patients diagnosed or under investigation for COVID-19. A machine learning algorithm was trained to predict future presence of intubation based on prior vitals, laboratory, and demographic data. Model performance was compared to ROX index, a validated prognostic tool for prediction of mechanical ventilation. RESULTS: 4087 patients admitted to five hospitals between February 2020 and April 2020 were included. 11.03% of patients were intubated. The machine learning model outperformed the ROX-index, demonstrating an area under the receiver characteristic curve (AUC) of 0.84 and 0.64, and area under the precision-recall curve (AUPRC) of 0.30 and 0.13, respectively. In the Kaplan-Meier analysis, patients alerted by the model were more likely to require intubation during their admission (p < 0.0001). CONCLUSION: In patients diagnosed or under investigation for COVID-19, machine learning can be used to predict future risk of intubation based on clinical data which are routinely collected and available in clinical setting. Such an approach may facilitate identification of high-risk patients to assist in clinical care.

Deep Learning Automates Measurement of Spinopelvic Parameters on Lateral Lumbar Radiographs.

STUDY DESIGN: Cross-sectional database study. OBJECTIVE: The objective of this study was to develop an algorithm for the automated measurement of spinopelvic parameters on lateral lumbar radiographs with comparable accuracy to surgeons. SUMMARY OF BACKGROUND DATA: Sagittal alignment measurements are important for the evaluation of spinal disorders. Manual measurement methods are time-consuming and subject to rater-dependent error. Thus, a need exists to develop automated methods for obtaining sagittal measurements. Previous studies of automated measurement have been limited in accuracy, inapplicable to common plain films, or unable to measure pelvic parameters. METHODS: Images from 816 patients receiving lateral lumbar radiographs were collected sequentially and used to develop a convolutional neural network (CNN) segmentation algorithm. A total of 653 (80%) of these radiographs were used to train and validate the CNN. This CNN was combined with a computer vision algorithm to create a pipeline for the fully automated measurement of spinopelvic parameters from lateral lumbar radiographs. The remaining 163 (20%) of radiographs were used to test this pipeline. Forty radiographs were selected from the test set and manually measured by three surgeons for comparison. RESULTS: The CNN achieved an area under the receiver-operating curve of 0.956. Algorithm measurements of L1-S1 cobb angle, pelvic incidence, pelvic tilt, and sacral slope were not significantly different from surgeon measurement. In comparison to criterion standard measurement, the algorithm performed with a similar mean absolute difference to spine surgeons for L1-S1 Cobb angle (4.30°â€Š±â€Š4.14° vs. 4.99°â€Š±â€Š5.34°), pelvic tilt (2.14°â€Š±â€Š6.29° vs. 1.58°â€Š±â€Š5.97°), pelvic incidence (4.56°â€Š±â€Š5.40° vs. 3.74°â€Š±â€Š2.89°), and sacral slope (4.76°â€Š±â€Š6.93° vs. 4.75°â€Š±â€Š5.71°). CONCLUSION: This algorithm measures spinopelvic parameters on lateral lumbar radiographs with comparable accuracy to surgeons. The algorithm could be used to streamline clinical workflow or perform large scale studies of spinopelvic parameters.Level of Evidence: 3.