Non-home Discharge Disposition Research Articles

The association between surgical start time and spine surgery outcomes

ObjectiveNeurosurgical operations, including spine surgeries, often occur outside “normal business hours” due to the urgent or emergent nature of cases. This study investigates the association of surgical start time (SST) with spine surgery outcomes. MethodsA retrospective cross-sectional study was performed using electronic health record data from a multi-hospital academic health system from 2017 to 2024. Eligible patients included adults who underwent spine surgery with a recorded SST. Patients were separated into a regular hours group (7:00 A.M. to 5:00 P.M.) and an afterhours group (SST outside this time window). The association between SST and extended length of stay (greater than 3 days), readmission, and discharge disposition was examined. ResultsThe sample included 12,658 patients with 10,737 (84.8 %) patients in the regular hours group and 1921 (15.2 %) patients in the afterhours group. Afterhours SST had significantly increased rates of extended length of stay, non-home discharge disposition, and readmission compared to regular hours SST. Adjusting for age, comorbidities, case classification, the time from admission to SST, and surgery type, afterhours SST was significantly associated with non-home discharge disposition (OR 1.27, 95 % CI 1.12 – 1.45, p < 0.001). ConclusionThis is the largest study to examine the association of SST with outcomes of spine surgery. Controlling for potential confounders, afterhours SST was significantly associated with non-home discharge disposition.

Characterizing a Common Phenomenon: Why do Trauma Patients Re-present to the Emergency Department?

IntroductionTrauma patients return to the emergency department (ED) at alarmingly high rates, despite not all patients requiring hospital resources. Reasons for ED re-presentation and associated risk factors have not been fully investigated. MethodsRetrospective cohort study of adult trauma admissions at an urban safety net level 1 trauma center (1/12018-12/312021). Risk factors for ED re-presentation were identified using purposeful selection and modeled using multivariable logistic regression. ResultsOf 2491 patients, 19% returned within 30 d (N = 475). Most patients presented for uncontrolled pain (37%, N = 175), medical concerns (25%, N = 119), and infection (10%, N = 49). The readmission rates varied as follows: 18% for uncontrolled pain (N = 32), 42% for medical concerns (N = 50), and 67% for infection (N = 33). Risk factors for uncontrolled pain included depression/anxiety (adjusted odds ratio [aOR] 2.06, 95% confidence interval [CI] 1.39-3.05), substance use disorder (SUD) (aOR 1.65, 95% CI 1.12-2.43), and penetrating mechanism of injury (aOR 2.25, 95% CI 1.59-3.18). Risk factors for medical concerns included number of medical comorbidities (aOR 1.34, 95% CI 1.18-1.52), depression/anxiety (aOR 1.97, 95% CI 1.28-3.01), SUD (aOR 2.48, 95% CI 1.65-3.74), and nonhome discharge disposition (aOR 1.56, 95% CI 1.07-2.28). Risk factors for infection included non-English primary language (aOR 3.41, 95% CI 1.82-6.39), SUD (aOR 2.00, 95% CI 1.03-3.88), and nonhome discharge disposition (aOR 2.06, 95% CI 1.15-3.67). ConclusionsUncontrolled pain was the most common reason for re-presentation, although only a small fraction required readmission. Patients with penetrating injury may benefit from improved pain control. Primary care provider follow-up may help mitigate risk of medical disease exacerbation, and wound care instructions for non–English speaking patients may decrease re-presentation for infection.

The Utility of Neighborhood Social Vulnerability Indices in Predicting Non-Home Discharge Disposition Following Revision Total Joint Arthroplasty: A Comparison Study

BackgroundIdentifying risk factors associated with non-home discharge (NHD) following revision hip and knee total joint arthroplasty (TJA) could reduce the rate of preventable discharge to rehabilitation or skilled nursing facilities. Neighborhood-level deprivation indices are becoming an increasingly important measure of socioeconomic disadvantage as these indices consider multiple social determinants of health. This study aimed to compare the utility of widely used neighborhood social vulnerability indices in predicting NHD following revision TJA patients. MethodsThis study included 1,043 consecutive patients who underwent revision TJA at a single tertiary health system. There were three multivariate logistic regression analyses with the outcome of NHD performed using the area deprivation index (ADI), social deprivation index (SDI), and social vulnerability index while controlling for other demographic variables. Neighborhood-level indices were included in the analysis as continuous variables and categorical quartiles, with the lowest quartile representing the least deprived neighborhoods of the patient cohort. The strength of the association of significant indices was measured. ResultsPatients in the highest ADI and SDI quartiles demonstrated higher odds of NHD compared to the cohort with the lowest quartile (ADI OR [odds ratio] = 1.93, 95% CI [confidence interval] = 1.23 to 3.03, P = 0.005; SDI OR = 1.86, 95% CI = 1.18 to 2.91, P = 0.007). Discharge disposition was more strongly associated with ADI than SDI (0.68 versus 0.26). Age, American Society of Anesthesiologist status, and alcohol use were independent determinants of discharge disposition. No significant association was seen between social vulnerability index and discharge disposition. ConclusionsArea-level indices can be utilized to identify patients at higher risk of NHD following revision TJA. This study highlights the important differences between these indices' utility when evaluating their effects on clinical outcomes in this patient population. The findings shed light on the potential of integrating these tools into policy development, clinical preoperative programs, and research to better understand and address the health disparities in arthroplasty outcomes.

Neighborhood Socioeconomic Disadvantage is Associated with Increased Healthcare Utilization After Septic and Aseptic Revision Total Hip Arthroplasty

BackgroundA greater Area Deprivation Index (ADI), a tool that gauges socioeconomic disadvantage at the neighborhood level, is associated with worse healthcare outcomes following primary total hip arthroplasty (THA). However, its association with revision THA (rTHA) is unknown. This study aimed to determine the association between ADI and rates of postoperative healthcare resource utilization following rTHA. MethodsA total of 996 patients who underwent rTHA between 2016 and 2022 were enrolled in a prospective study. The primary outcomes assessed were non-home discharge disposition (DD), length of stay (LOS) ≥ 3 days, 90-day emergency department (ED) visits, and 90-day hospital readmissions. The ADI was calculated using the patient’s home address at the time of surgery, with greater ADI indicating greater socioeconomic disadvantage. We evaluated the mediation effect of patient race on ADI and postoperative healthcare utilization using a multivariable logistic regression model. ResultsA higher median ADI was revealed for patients who experienced non -home discharge (P = 0.001), extended LOS (P < 0.001), and ED readmission within 90-days of surgery (P = 0.045). When comparing septic versus aseptic rTHA patients, there were significant differences in healthcare resource utilization but no difference in ADI between the two groups. For aseptic rTHA, ADI significantly mediated the effect of race on both non-home DD and LOS ≥ 3 (41% and 46% mediation, respectively). In septic rTHA, ADI mediated 31.1% of the effect of race on non-home DD, but showed minimal mediation effect on LOS. The mediation effect of ADI on ED admission and hospital readmission was minimal for both groups ConclusionHigher ADI scores are associated with increased healthcare utilization after rTHA, including longer hospital stays and more non-home discharges. The ADI significantly mediates the effect of race on these outcomes, particularly in aseptic rTHA cases, suggesting that neighborhood socioeconomic factors play a crucial role in previously observed racial disparities.

Frailty, Myelopathy, and Enhanced Recovery after Surgery in Patients Undergoing Posterior Cervical Fusion

BackgroundThe Enhanced Recovery After Surgery (ERAS) protocol is a proven method to improve postsurgical outcomes. While recent studies have shown the benefit of ERAS even in frail patient populations, myelopathy is another factor affecting outcomes in patients undergoing Posterior Cervical Fusion (PCF). This study evaluated the benefit of an ERAS protocol in frail patients undergoing PCF. MethodsA retrospective chart review identified consecutive patients undergoing PCF by a single surgeon from August 2015-July 2021, with implementation of ERAS in December 2018. Outcome measures included length of stay (LOS), non-home discharge disposition, complications, return of physiologic function, and severe pain score. A mFI-5 score of ≥ 2 and a Nurick score of ≥ 3 defined frail and myelopathic patients, respectively. Univariate analysis (p<0.05) and multivariate analyses, using mixed-effect models, (p<0.0125), were performed. ResultsThere were a total of 174 patients, 71 frail (41%). Of the frail patients, 61% were also myelopathic and 56% underwent ERAS. Of the non-frail patients, 43% were myelopathic and 57% underwent ERAS. On univariate analyses, frail patients with ERAS had less drains placed (p<0.0001), decreased urinary retention (p=0.0002), decreased LOS (p=0.013), and were less likely to have a non-home discharge (p=0.001). On multivariate analysis, LOS (p=0.0003), time to return of physiologic function (p=0.004), complications (p=0.001), and non-home discharges (p<0.0001) were decreased with ERAS, irrespective of groups. ConclusionsERAS is an effective protocol in PCF patients that may expedite return of physiologic function, lessen LOS, decrease the number of non-home discharges, and reduce complications, irrespective of frailty or myelopathy status.

Atrial fibrillation and short-term outcomes after cancer-related ischemic stroke.

Atrial fibrillation (AF) and cancer are each associated with worse outcomes in patients with acute ischemic stroke (AIS). Few studies have evaluated the impact of AF on outcomes of cancer-related stroke. We conducted a retrospective cross-sectional study using the 2016-2019 National Inpatient Sample, identifying all hospitalizations with diagnosis codes for cancer and AIS. The primary exposure was a diagnosis of AF. The primary outcome was in-hospital mortality. The secondary outcomes were length-of-stay and discharge to non-home locations. We used multiple logistic and linear regression models, adjusted for age, gender, race-ethnicity, and the Charlson Comorbidity Index, to examine the association between AF and study outcomes. Among 150,200 hospitalizations with diagnoses of cancer and AIS (mean age 72 years, 53% male), 40,084 (26.7%) included comorbid AF. Compared to hospitalizations without AF, hospitalizations with AF had higher rates of in-hospital mortality (14.8% [95% CI, 14.0%-15.6%] vs 12.1% [95% CI, 11.6%-12.5%]) and non-home discharge disposition (83.5% [95% CI, 82.7%-84.3%] vs 75.1% [95% CI, 74.5%-75.7%]) as well as longer mean length-of-stay (8.4 days [95% CI, 8.2-8.6 days] vs 8.2 days [95% CI, 8.0-8.3 days]). In multivariable analyses, AF remained independently associated with higher odds of in-hospital mortality (adjusted odds ratio [aOR], 1.34; 95% CI, 1.24-1.46), non-home discharge disposition (aOR, 1.32; 95% CI, 1.23-1.42), and longer length-of-stay (adjusted mean difference, 13.7%; 95% CI, 10.9%-16.7%). In cancer-related AIS, comorbid AF is associated with worse short-term outcomes, including higher odds for in-hospital mortality, poor discharge disposition, and longer hospital stays.

The association of ASA score and outcomes following to.tal ankle arthroplasty

BackgroundThis study seeks to evaluate the relationship between American Society of Anesthesiologist (ASA) score and postoperative outcomes following TAA. MethodsThe American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database was queried from 2007 to 2020 to identify 2210 TAA patients. Patients were stratified into low (n = 1328; healthy/mild systemic disease) or high (n = 881; severe/life-threatening systemic disease) ASA score cohorts. ResultsThere was no statistically significant difference in complications, readmission, or reoperation rate based on ASA score. Increased ASA score was significantly associated with longer length of stay (low = 1.69 days, high = 1.98 days; p < .001) and higher rate of adverse discharge (low = 95.3 %, high = 87.4 %; p < .001). ConclusionHigher ASA scores (3 and 4) were statically significantly associated with increased length of stay and non-home discharge disposition. These findings are valuable for physicians and patients to consider prior to TAA given the increased utilization of resources and cost associated with higher ASA scores. Level of EvidenceLevel III, Retrospective cohort study

Abstract TP230: Atrial Fibrillation and Short-Term Outcomes After Cancer-Related Ischemic Stroke

Introduction: Atrial fibrillation (AF) and cancer are both independently associated with worse outcomes in patients with acute ischemic stroke. Few studies have evaluated the impact of AF on outcomes of cancer-related stroke. We aimed to determine if AF is associated with worse short-term outcomes in patients with stroke and comorbid cancer. Methods: Using the 2016-2019 National Inpatient Sample, we identified all hospitalizations with diagnosis codes for cancer and ischemic stroke. The primary exposure was a diagnosis of AF. The primary outcome was in-hospital mortality. The secondary outcomes were length-of-stay and discharge other than to home. We used multivariable logistic and linear regression models, as appropriate, adjusted for age, gender, race-ethnicity, and the Charlson Comorbidity Index, to examine the association between AF and study outcomes. Results: Among 150,200 hospitalizations with diagnoses of cancer and ischemic stroke (mean age 72 years, 53% male), 40,084 (26.7%) included comorbid AF. Compared to hospitalizations without AF, hospitalizations with AF had higher rates of in-hospital mortality (14.8% vs 12.1%, p&lt;0.001) and non-home discharge disposition (83.5% vs 75.1%, p&lt;0.001) as well as longer mean length-of-stay (8.4 days vs 8.2 days, p&lt;0.001). In multivariable analyses, AF remained independently associated with higher odds of in-hospital mortality (adjusted OR, 1.34; 95% CI, 1.24-1.46), non-home discharge disposition (adjusted OR, 1.32; 95% CI, 1.23-1.42), and longer length-of-stay (adjusted mean difference, 13.7%; 95% CI, 10.9%-16.7%). Conclusions: In cancer-related ischemic stroke, comorbid AF is associated with worse short-term outcomes, including higher odds for in-hospital mortality, worse discharge disposition, and longer hospital stays.

The Effect of ASA Score on Total Ankle Arthroplasty

Category: Ankle; Other Introduction/Purpose: The American Society of Anesthesiologist (ASA) physical status classification system serves as a useful measure of patient comorbidities. Multiple studies have found an increased risk of readmission in patients undergoing total knee or hip arthroplasty with elevated ASA scores. There is a paucity of literature investigating the association of ASA score and postoperative outcomes following total ankle arthroplasty (TAA). Thus, this study seeks to evaluate the relationship between ASA score and postoperative outcomes following TAA. Methods: The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) database was queried from 2007-2020 to identify 2210 TAA patients. Patients were stratified based on preoperative ASA score, 1-normal healthy patient (n=101), 2-mild systemic disease (n=1228), 3-severe systemic disease (n=855), 4-severe, life-threatening systemic disease (n=26). Demographics, medical comorbidities, concomitant procedures, hospital length of stay (LOS), and 30-day complication, readmission, and reoperation rates were compared between groups. The cohort was predominantly male (54.0%) and mean patient age was 63.99 (range 19-87) years with a mean BMI of 31.01 (range 17.14-57.78) m/kg2. Results: Patients with an ASA score ≥3 were significantly older (≤2 =62.78, ≥3 =65.81%; p&lt;.001), with higher BMI (≤2 =29.72, ≥3=32.97; p&lt;.001). 2210 patients were discharged following TAA, with 37 readmitted within 30 days following surgery. There was no statistically significant difference in readmission or reoperation rate based on ASA score. (Table 1) The average LOS following TAA was 1.81 (range 0-28) days, and increased ASA score was statistically significantly associated with longer length of stay (≤2=1.69 days, ≥3 =1.98 days; p&lt;.001). Patients with ASA score ≥3 were discharged home at a significantly lower rate than those with lower ASA scores (≤2 =95.3%, ≥3 =87.4%; p&lt;.001). There was no statistically significant difference in postoperative complication rates within 30 days of surgery. Conclusion: Higher ASA score was statically significantly associated with increased length of stay and non-home discharge disposition. These findings are valuable for physicians and patients to consider prior to TAA given the increased utilization of resources and cost associated with higher ASA scores.

Severe Obesity Is an Independent Risk Factor of Early Readmission and Nonhome Discharge After Cervical Disc Replacement.

Despite growing interest in cervical disc replacement (CDR) for conditions such as cervical radiculopathy, limited data exists describing the impact of obesity on early postoperative outcomes and complications. These data are especially important as nearly half of the adult population in the United States is expected to become obese (body mass index [BMI] ≥ 30 kg/m2) by 2030. The goal of this study was to compare the demographics, perioperative variables, and complication rates following CDR. The 2005-2020 American College of Surgeons National Surgical Quality Improvement Program datasets were queried for patients who underwent primary 1- or 2-level CDR. Patients were divided into 3 cohorts: Nonobese (BMI: 18.5-29.9 kg/m2), Obese class-I (BMI: 30-34.9 kg/m2), Obese class-II/III (BMI ≥ 35 kg/m2). Morbidity was defined as the presence of any complication within 30 days postoperatively. Rates of 30-day readmission, reoperation, morbidity, individual complications, length of stay, frequency of nonhome discharge disposition were collected. A total of 5,397 patients were included for analysis: 3,130 were nonobese, 1,348 were obese class I, and 919 were obese class II/III. There were more 2-level CDRs performed in the class II/III cohort compared to the nonobese group (25.7% vs. 21.5%, respectively; p < 0.05). Class-II/III had more nonhome discharges than class I and nonobese (2.1% vs. 0.5% vs. 0.7%, respectively; p < 0.001). Readmission rates differed as well (nonobese: 0.5%, class I: 1.1%, class II/III: 2.1%; p < 0.001) with pairwise significance between class II/II and nonobese. Class II/III obesity was an independent risk factor for both readmission (odds ratio [OR], 3.32; p = 0.002) and nonhome discharge (OR, 2.51; p = 0.02). Neither 30-day reoperation nor morbidity rates demonstrated significance. No mortalities were reported. Although obese class-II/III were risk factors for 30-day readmission and nonhome discharge, there was no significant difference in reoperation rates or morbidity. CDR procedures can continue to be safely preformed independent of obesity status.

344 Drivers of Adverse 30-Day Outcomes for Pediatric Spine Tumor Surgery Patients: A Nationwide Analysis

INTRODUCTION: Spine tumors in children are an uncommon clinical entity. Drivers of reoperation and readmission following surgery for this pathology are poorly characterized. METHODS: We conducted a retrospective cohort analysis of all spine tumor surgery cases in the National Surgical Quality Improvement Program-Pediatrics database from 2012-2020, using diagnosis and Current Procedural Terminology codes. The primary outcome was a composite of 30-day mortality, reoperation, readmission, and non-home discharge disposition. Stepwise regression with bidirectional elimination via the Akaike Information Criterion was used to identify predictors of adverse 30-day outcomes. 25 candidate variables were assessed, including tumor characteristics, patient demographics, preoperative comorbidities, operative details, and postoperative complications. RESULTS: The analytic cohort included 728 pediatric patients undergoing spine tumor surgery. Patients predominantly had intradural intramedullary tumors (79.3%). Average age was 7.7 years and 54% were male. 86.5% were discharged to home. 30-day mortality rate was 0.4%. 48 patients (6.6%) required reoperation, with the most common reasons being infection (n = 14), repair of cerebrospinal fluid leak (n = 8), shunt placement (n = 8), and tumor recurrence (n = 6). 39 patients (5.4%) underwent readmission within 30 days, primarily for surgical site infection (n = 11), nervous system complication (n = 8), or wound dehiscence (n = 7). Stepwise regression identified predictors of adverse 30-day outcomes including emergent surgery (OR = 2.64, P = 0.014), operative time (OR = 1.03 per +10 minutes, P &lt; 0.001), length of stay (OR = 1.03 per +1 day, P = 0.035), surgical site infection or dehiscence (OR = 5.26, P &lt; 0.001), nervous system complications (OR = 10.31, P = 0.009), and non-surgical site infections (OR = 27.95, P &lt; 0.001). When mortality, reoperation, and readmission were analyzed in aggregate, emergent surgery and postoperative complications remained as significant predictors. CONCLUSIONS: 30-day adverse events following pediatric spine tumor surgery were predominantly driven by emergent surgery, prolonged operative or hospitalization time, and postoperative complications.

322 Distributed Vector Representations of Neurosurgical Admission Notes Predict Discharge Disposition

INTRODUCTION: Natural language processing (NLP) aims to extract information from unstructured language input for use in machine learning, without the need for laborious manual review of text to retrieve data of potential interest. An NLP-based model utilizing term frequency-inverse document frequency (TF-IDF) vectoral representations of pre-operative clinical notes has previously been shown to predict discharge disposition following meningioma resection. The TF-IDF vectorization scheme is based on word counts across documents but does not account for word order or semantic information. An alternative algorithm, paragraph vectorization, utilizes an unsupervised neural network to learn document vector representations. METHODS: We analyzed 7,122 admission notes. We trained a neural network to derive paragraph vector representations of notes and trained a classifier to predict home vs non-home discharge disposition based on these representations. Out-of-sample performance was assessed on a held-out testing set. We repeated the analysis using TF-IDF document representations to compare the two approaches. Performance was quantified using area under the receiver operating characteristic curve (ROC-AUC). RESULTS: Model performance increased with vector dimensionality. A 200-dimensional distributed document vectorization was highly predictive of nonhome discharge with an ROC-AUC of 0.82 ± 0.01 (mean ± standard error over 5-fold nested cross-validation). When the dimensionality of the embedding vector space was relatively low (e.g. &lt;100), learned paragraph vectors predicted discharge disposition better than TF-IDF vectors. CONCLUSIONS: Natural language processing of admission notes via neural network-learned distributed vectoral representations is a viable strategy of information extraction from neurosurgical clinical documentation. This embedding scheme overcomes some of the limitations of TF-IDF and encodes more information in lower-dimensional embedding vector spaces. It holds promise as a method of using free text to train machine learning algorithms for prediction of clinically relevant outcomes.

Endoscopic and Nonendoscopic Approaches to Single-Level Lumbar Spine Decompression: Propensity Score-Matched Comparative Analysis and Frailty-Driven Predictive Model.

The endoscopic spine surgery (ESS) approach is associated with high levels of patient satisfaction, shorter recovery time, and reduced complications. The present study reports multicenter, international data, comparing ESS and non-ESS approaches for singlelevel lumbar decompression, and proposes a frailty-driven predictive model for nonhome discharge (NHD) disposition. Cases of ESS and non-ESS lumbar spine decompression were queried from the American College of Surgeons National Surgical Quality Improvement Program database (2017-2020). Propensity score matching was performed on baseline characteristics frailty score (measured by risk analysis index [RAI] and modified frailty index-5 [mFI-5]). The primary outcome of interest was NHD disposition. A predictive model was built using logistic regression with RAI as the primary driver. Single-level nonfusion spine lumbar decompression surgery was performed in 38,686 patients. Frailty, as measured by RAI, was a reliable predictor of NHD with excellent discriminatory accuracy in receiver operating characteristic (ROC) curve analysis: C-statistic: 0.80 (95% confidence interval [CI], 0.65-0.94) in ESS cohort, C-statistic: 0.75 (95% CI, 0.73-0.76) overall cohort. After propensity score matching, there was a reduction in total operative time (89 minutes vs. 103 minutes, p = 0.049) and hospital length of stay (LOS) (0.82 days vs. 1.37 days, p < 0.001) in patients treated endoscopically. In ROC curve analysis, the frailty-driven predictive model performed with excellent diagnostic accuracy for the primary outcome of NHD (C-statistic: 0.87; 95% CI, 0.85-0.88). After frailty-based propensity matching, ESS is associated with reduced operative time, shorter hospital LOS, and decreased NHD. The RAI frailty-driven model predicts NHD with excellent diagnostic accuracy and may be applied to preoperative decisionmaking with a user-friendly calculator: nsgyfrailtyoutcomeslab.shinyapps.io/lumbar_decompression_dischargedispo.

Frailty Measured by the Risk Analysis Index Predicts Nonhome Discharge and Mortality After Resection in Refractory Epilepsy: Analysis of 1236 Patients From a Prospective Surgical Registry, 2012 to 2020.

Risk stratification of epilepsy surgery patients remains difficult. The Risk Analysis Index (RAI) is a frailty measurement that augments preoperative risk stratification. To evaluate RAI's discriminative threshold for nonhome discharge disposition (NHD) and mortality (or discharge to hospice within 30 days of operation) in epilepsy surgery patients. Patients were queried from the American College of Surgeons-National Surgical Quality Improvement Program database (2012-2020) using diagnosis/procedure codes. Linear-by-linear trend tests assessed RAI's relationship with NHD and mortality. Discriminatory accuracy was assessed by C-statistics (95% CI) in receiver operating characteristic curve analysis. Epilepsy resections (N = 1236) were grouped into temporal lobe (60.4%, N = 747) and nontemporal lobe (39.6%, N = 489) procedures. Patients were stratified by RAI tier: 76.5% robust (RAI 0-20), 16.2% normal (RAI 21-30), 6.6% frail (RAI 31-40), and 0.8% severely frail (RAI 41 and above). The NHD rate was 18.0% (N = 222) and positively associated with increasing RAI tier: 12.5% robust, 34.0% normal, 38.3% frail, and 50.0% severely frail ( P < .001). RAI had robust predictive discrimination for NHD in overall cohort (C-statistic 0.71), temporal lobe (C-statistic 0.70), and nontemporal lobe (C-statistic 0.71) cohorts. The mortality rate was 2.7% (N = 33) and significantly associated with RAI frailty: 1.1% robust, 8.0% normal, 6.2% frail, and 20.0% severely frail ( P < .001). RAI had excellent predictive discrimination for mortality in overall cohort (C-statistic 0.78), temporal lobe (C-statistic 0.80), and nontemporal lobe (C-statistic 0.74) cohorts. The RAI frailty score predicts mortality and NHD after epilepsy surgery. This is accomplished with a user-friendly calculator: https://nsgyfrailtyoutcomeslab.shinyapps.io/epilepsy/ .

Internal and External Validation of the Generalizability of Machine Learning Algorithms in Predicting Non-home Discharge Disposition Following Primary Total Knee Joint Arthroplasty

BackgroundNonhome discharge disposition following primary total knee arthroplasty (TKA) is associated with a higher rate of complications and constitutes a socioeconomic burden on the health care system. While existing algorithms predicting nonhome discharge disposition varied in degrees of mathematical complexity and prediction power, their capacity to generalize predictions beyond the development dataset remains limited. Therefore, this study aimed to establish the machine learning model generalizability by performing internal and external validations using nation-scale and institutional cohorts, respectively. MethodsFour machine learning models were trained using the national cohort. Recursive feature elimination and hyper-parameter tuning were applied. Internal validation was achieved through five-fold cross-validation during model training. The trained models’ performance was externally validated using the institutional cohort and assessed by discrimination, calibration, and clinical utility. ResultsThe national (424,354 patients) and institutional (10,196 patients) cohorts had non-home discharge rates of 19.4 and 36.4%, respectively. The areas under the receiver operating curve of the model predictions were 0.83 to 0.84 during internal validation and increased to 0.88 to 0.89 during external validation. Artificial neural network and histogram-based gradient boosting elicited the best performance with a mean area under the receiver operating curve of 0.89, calibration slope of 1.39, and Brier score of 0.14, which indicated that the two models were robust in distinguishing non-home discharge and well-calibrated with accurate predictions of the probabilities. The low inter-dataset similarity indicated reliable external validation. Length of stay, age, body mass index, and sex were the strongest predictors of discharge destination after primary TKA. ConclusionThe machine learning models demonstrated excellent predictive performance during both internal and external validations, supporting their generalizability across different patient cohorts and potential applicability in the clinical workflow.

Racial Disparities in Outcomes After THA and TKA Are Substantially Mediated by Socioeconomic Disadvantage Both in Black and White Patients.

Racial Disparities in Outcomes After THA and TKA Are Substantially Mediated by Socioeconomic Disadvantage Both in Black and White Patients.

P132. Lumbar fusion during the COVID-19 pandemic: greater rates of morbidity and longer procedures

BACKGROUND CONTEXTThe coronavirus-19 (COVID-19) pandemic has altered the standard of care for spine surgery in many ways. However, there is a lack of literature evaluating the potential changes in surgical outcomes and perioperative factors for spine procedures performed during the pandemic. In particular, no large database study evaluating the impact of the COVID-19 pandemic on spine surgery outcomes has been published.PURPOSETo evaluate the impact of the COVID-19 pandemic on perioperative factors and postoperative outcomes of lumbar fusion procedures.STUDY DESIGN/SETTINGRetrospective cohort study.PATIENT SAMPLEPatients who underwent primary, elective lumbar fusionbetween 2019 and 2020.OUTCOME MEASURESThirty-day readmission, reoperation, morbidity, complications, operative time, non-home discharge disposition, length of hospital stay, total RVUs and RVUs per minute.METHODSThe ACS-NSQIP database was queried for all patients who underwent primary lumbar fusion in 2019 and 2020. Exclusion criteria included cervical or thoracic fusion, arthroplasty, non-elective, emergency cases, deformity procedures, malignancy, preoperative sepsis, wound infection and missing perioperative variables. Patients were grouped into cohorts based on 2019 or 2020 operation year. Multivariate logistic and Poisson regression models were utilized to evaluate the impact of 2020 operative year on postoperative outcomes.RESULTSA total of 27,446 patients were included in the analysis (12,473 cases in 2020). For the 2020 cohort, non-white race (24.9% vs 23.4%, p=0.004), functional dependence (1.8% vs 1.4%, p=0.022), congestive heart failure (0.5% vs 0.3%, p=0.015), and ASA class =3 (53.5% vs 52.2%, p=0.037) were more common. Lumbar fusions performed in 2020 were also associated with longer mean operative time (209.3±99.3 vs. 205.9±98.6 minutes, p=0.005) and higher mean fusion levels (1.45±0.69 vs 1.41±0.65, p 0.05). However, operation year 2020 independently predicted morbidity (OR 1.087 [95% CI 1.007-1.172], p=0.032), pneumonia (1.407 [1.039-1.906], p=0.027), DVT (1.508 [1.106-2.056], p=0.009) and sepsis (1.537 [1.101-2.146], p=0.012). In terms of perioperative variables, operation year 2020 predicted greater operative time (1.012 [1.001-1.013], p=0.046), non-home discharge (0.690 [0.642-0.743], p<0.001), and total RVUs (1.044 [1.004-1.083], p=0.033).CONCLUSIONSLumbar fusion procedures performed amidst the COVID-19 pandemic were associated with poorer outcomes, including higher rates of morbidity, pneumonia, DVT and sepsis. In addition, surgeries performed in 2020 were associated with longer operative times and less frequent non-home discharge disposition.FDA DEVICE/DRUG STATUSThis abstract does not discuss or include any applicable devices or drugs.

AM-PAC mobility scores predict non-home discharge following adult spinal deformity surgery

BACKGROUND CONTEXTAdult spinal deformity (ASD) surgery requires an extended recovery period and often non-routine discharge. The Activity Measure for Post-Acute Care (AM-PAC) Basic Mobility Inpatient Short Form (6-Clicks) is a prediction tool, validated for other orthopedic procedures, to assess a patient's ability to mobilize after surgery. PURPOSETo assess the thresholds of AM-PAC scores that determine non-home discharge disposition in patients who have undergone ASD surgery. STUDY DESIGNRetrospective review PATIENT SAMPLENinety consecutive ASD patients with ≥5 levels fused who underwent surgery from 2015 to 2018, with postoperative AM-PAC scores measured before discharge, were included. OUTCOME MEASURESNon-home discharge disposition METHODSPatients with routine home discharge were compared to those with non-home discharge. Bivariate analysis was first conducted to compare these groups by preoperative demographics, comorbidities, radiographic alignment, surgical characteristics, HRQOLs, and AM-PAC measurements. Threshold linear regression with Bayesian information criteria was utilized to identify the optimal cutoffs for AM-PAC scores associated with increased likelihood of non-home discharge. Finally, multivariable analysis controlling for age, sex, comorbidities, levels fused, perioperative complication, and home support was conducted to assess each threshold. RESULTSThirty-six (40%) of 90 patients analyzed had non-home discharge. On bivariate analysis, first AM-PAC score (13.5 vs. 17), last AM-PAC score (17 vs. 20), and AM-PAC change per day (+.387 vs. +1) were all significantly associated with non-home discharge. Threshold regression identified that cutoffs of ≤15 for first AM-PAC score, <17 for last AM-PAC score, and <+0.625 for daily AM-PAC change were associated with non-home discharge. On multivariable analysis, first AM-PAC score ≤15 (odds ratio [OR] 11.28; confidence interval [CI] 2.96-42.99; p<.001), last AM-PAC score <17 (OR 33.57; CI 5.85-192.82; p<.001), and AM-PAC change per day <+0.625 (OR 6.24; CI 2.01-19.43; p<.001) were all associated with increased odds of non-home discharge. CONCLUSIONSFirst AM-PAC score of 15 or less can help predict non-home discharge. A goal of daily AM-PAC increases of 0.625 points toward a final AM-PAC score of 17 can aid in achieving home discharge. The early AM-PAC mobility threshold of ≤15 may help prepare for non-home discharge, while AM-PAC daily changes per day <0.625 and final AM-PAC <17 may provide goals for mobility improvement during the early postoperative period in order to prevent non-home discharge.

Preoperative Veterans Rand-12 Mental Composite Score of < 40 Leads to Increased Healthcare Utilization and Diminished Improvement After Primary Knee Arthroplasty

BackgroundAdverse outcomes after total knee arthroplasty (TKA) have been associated with preoperative psychological disorders and poor mental health. We aimed to investigate and quantify the association between preoperative mental health and 1) postoperative 90-day health care utilization; and 2) 1-year patient-reported outcomes after primary TKA. MethodsRetrospective review of prospectively collected data of patients who underwent primary elective TKA (n = 7,476) was performed. Preoperative mental health was evaluated using Veterans Rand-12 Mental Composite Scores (VR-12 MCS). Outcomes included prolonged length of stay (>2-days), nonhome discharge, 90-day readmissions, emergency department visits, and reoperation. Improvement in Knee Injury and Osteoarthritis Outcome Score (KOOS) and Patient Acceptable Symptom State (PASS) achievement were evaluated at 1-year. Multivariable regression was implemented to explore associations between preoperative VR-12 MCS and outcomes of interest. ResultsA total of 5,402 (72.3%) completed 1-year follow-up. Lower preoperative VR-12 MCS was associated with higher odds of prolonged length of stay (MCS 20-39: odds ratio (OR): 1.46;P < .001), and nonhome discharge disposition (MCS 20-39: OR: 1.92;P < .001), but not 90-day readmission or reoperation (MCS20-39; P = .12 and P = .64). At 1-year, patients with a lower MCS were less likely to attain a substantial clinical benefit in KOOS-pain (MCS 0-19; OR: 0.25; P < .001) and less likely to achieve PASS (MCS20-39; OR: 0.74; P = .002). Patients with an MCS >60 were more likely to be discharged home (OR: 1.42; P = .008), achieve substantial clinical benefit in their KOOS-JR (OR: 1.16; P = .027),-Pain (OR: 1.220; P = .007) and PASS at 1-year (OR: 1.28; P = .008). ConclusionsLower VR-12 MCS is associated with increased postoperative health care utilization and worse patient-reported outcome measures at 1-year post-TKA. These findings suggest that a VR-12 MCS ≤40 could be used to designate increased risk, guide the preoperative discussion and potential interventions.

Similar Healthcare Utilization and 1-Year Patient-Reported Outcomes between Cemented and Cementless Primary Total Knee Arthroplasty: A Propensity Score-Matched Analysis.

Cementless fixation for total knee arthroplasty (TKA) has gained traction with the advent of newer fixation technologies. This study assessed (1) healthcare utilization (length of stay (LOS), nonhome discharge, 90-day readmission, and 1-year reoperation); (2) 1-year mortality; and (3) 1-year joint-specific and global health-related patient-reported outcome measures (PROMs) among patients who received cementless versus cemented TKA. Patients who underwent cementless and cemented TKA at a single institution (July 2015-August 2018) were prospectively enrolled. A total of 424 cementless and 5,274 cemented TKAs were included. The cementless cohort was propensity score-matched to a group cemented TKAs (1:3-cementless: n = 424; cemented: n = 1,272). Within the matched cohorts, 76.9% (n = 326) cementless and 75.9% (n = 966) cementless TKAs completed 1-year PROMs. Healthcare utilization measures, mortality and the median 1-year change in knee injury and osteoarthritis outcome score (KOOS)-pain, KOOS-physical function short form (PS), KOOS-knee related quality of life (KRQOL), Veteran Rand (VR)-12 mental composite (MCS), and physical composite (PCS) scores were compared. The minimal clinically important difference (MCID) for PROMs was calculated. Cementless TKA exhibited similar rates of median LOS (p = 0.109), nonhome discharge disposition (p = 0.056), all-cause 90-day readmission (p = 0.226), 1-year reoperation (p = 0.597), and 1-year mortality (p = 0.861) when compared with cemented TKA. There was no significant difference in the median 1-year improvement in KOOS-pain (p = 0.370), KOOS-PS (p = 0.417), KOOS-KRQOL (p = 0.101), VR-12-PCS (p = 0.269), and VR-12-MCS (p = 0.191) between the cementless and cemented TKA cohorts. Rates of attaining MCID were similar in both cohorts for assessed PROMs (p > 0.05, each) except KOOS-KRQOL (cementless: n = 313 (96.0%) vs. cemented: n = 895 [92.7%]; p = 0.036). Cementless TKA provides similar healthcare-utilization, mortality, and 1-year PROM improvement versus cemented TKA. Cementless fixation in TKA may provide value through higher MCID improvement in quality of life. Future episode-of-care cost-analyses and longer-term survivorship investigations are warranted.