Adverse events in patients with high platelet reactivity following successful chronic total occlusion PCI: The Assessment of Dual AntiPlatelet Therapy with Drug-Eluting Stents (ADAPT-DES) study.

BACKGROUND: Chronic total occlusion (CTO) percutaneous coronary intervention (PCI) typically requires a greater number of stents and longer stent length than non-CTO PCI, placing these patients at greater risk for adverse ischemic events. We sought to determine whether the association between high platelet reactivity (HPR) and the risk of ischemic events is stronger after CTO than non-CTO PCI. METHODS: Patients undergoing successful PCI in the multicenter ADAPT-DES study were stratified according to whether they underwent PCI of a CTO. HPR was defined as VerifyNow platelet reaction units >208. The study primary endpoint was the 2-year risk target vessel failure ([TVF] defined as cardiac death, myocardial infarction, or target lesion revascularization). RESULTS: CTO PCI was performed in 400 of 8448 patients. HPR was present in 34.5% of CTO PCI patients and 43.1% of non-CTO PCI patients (P = .0007). Patients undergoing CTO PCI with versus without HPR had significantly higher 2-year rates of TVF (15.0% versus 8.3%, P = .04) without significant differences in bleeding. HPR was an independent predictor of 2-year TVF (adjusted HR 1.16, 95% CI 1.02-1.34, P = .03) whereas CTO PCI was not (adjusted HR 0.89, 95% CI 0.65-1.22, P = .48). There was a significant interaction between CTO versus non-CTO PCI and PRU as a continuous variable for 2-year TVF (Pinteraction = 0.02). CONCLUSIONS: In ADAPT-DES, HPR was associated with an increased 2-year risk of TVF after PCI, an association that was at least as strong after CTO PCI compared with non-CTO PCI.

Impact of CrossFit-Related Spinal Injuries.

INTRODUCTION: Exercise-related injuries (ERIs) are a common cause of nonfatal emergency department and hospital visits. CrossFit is a high-intensity workout regimen whose popularity has grown rapidly. However, ERIs due to CrossFit remained under investigated. METHODS: All patients who presented to the main hospital at a major academic center complaining of an injury sustained performing CrossFit between June 2010 and June 2016 were identified. Injuries were classified by anatomical location (eg, knee, spine). For patients with spinal injuries, data were collected including age, sex, body mass index (BMI), CrossFit experience level, symptom duration, type of symptoms, type of clinic presentation, cause of injury, objective neurological examination findings, imaging type, number of clinic visits, and treatments prescribed. RESULTS: Four hundred ninety-eight patients with 523 CrossFit-related injuries were identified. Spine injuries were the most common injuries identified, accounting for 20.9%. Among spine injuries, the most common location of injury was the lumbar spine (83.1%). Average symptom duration was 6.4 months ± 15.1, and radicular complaints were the most common symptom (53%). A total of 30 (32%) patients had positive findings on neurologic examination. Six patients (6.7%) required surgical intervention for treatment after failing an average of 9.66 months of conservative treatment. There was no difference in age, sex, BMI, or duration of symptoms of patients requiring surgery with those who did not. CONCLUSIONS: CrossFit is a popular, high-intensity style workout with the potential to injure its participants. Spine injuries were the most common type of injury observed and frequently required surgical intervention.

The safety of resection for primary central nervous system lymphoma: a single institution retrospective analysis.

Surgical resection is not the standard of care for primary central nervous system lymphoma (PCNSL), as historical studies have demonstrated unfavorable complication rates and limited benefits. Some recent studies suggest that resection may provide a therapeutic benefit, yet the safety of these procedures has not been systematically investigated in the setting of modern neurosurgery. We examined the safety of surgical resection for PCNSL. We retrospectively analyzed all patients with PCNSL treated at Columbia University Medical Center between 2000 and 2015 to assess complications rates following biopsy or resection using the Glioma Outcomes Project system. We identified predictors of complications and selection for resection. Well-validated scales were used to quantify patients' baseline clinical characteristics, including functional status, comorbid disease burden, and cardiac risk. The overall complication rate was 17.2% after resection, and 28.2% after biopsy. Cardiac risk (p = 0.047, OR 1.72 [1.01, 2.95]), and comorbid diagnoses (p = 0.004, OR 3.05 [1.42, 6.57]) predicted complications on multivariable regression. Patients who underwent resection had better KPS scores (median 70 v. 80, p = 0.0068, ∆ 10 [0.0, 10.00]), and were less likely to have multiple (46.5% v. 27.6%, p = 0.030, OR 1.42 [1.05, 1.92]) or deep lesions (70.4% v. 39.7%, p = 0.001, OR 1.83 [1.26, 2.65]). Age (p = 0.048, OR 0.75 per 10-year increase [0.56, 1.00]) and deep lesions (p = 0.003, OR 0.29 [0.13, 0.65]) influenced selection for resection on multivariable regression. Surgical resection of PCNSL is safe for select patients, with complication rates comparable to rates for other intracranial neoplasms. Whether there is a clinical benefit to resection cannot be concluded.

Chemoprophylactic Anticoagulation After Spine Surgery Does Not Increase Epidural Hematoma Risk: A Propensity-matched Analysis.

STUDY DESIGN: A retrospective cohort study. OBJECTIVE: To calculate the magnitude of any increased risk of epidural hematoma (EDH) associated with chemoprophylactic anticoagulation (chemoprophylaxis), if any. SUMMARY OF BACKGROUND DATA: Chemoprophylaxis for the prevention of venous thromboembolic events may be associated with an increased risk of EDH after spine surgery. MATERIALS AND METHODS: A total of 6869 consecutive spine surgeries performed at our institution were identified, and clinical and demographic data were collected. We identified cases in which symptomatic EDHs were evacuated within 30 days postoperatively. Patients receiving chemoprophylaxis and controls were matched using K-nearest neighbor propensity score matching to calculate the effect of anticoagulation on the rate of postoperative EDH. RESULTS: After propensity score matching, 1071 patients who received chemoprophylaxis were matched to 1585 controls. Propensity scores were well balanced between populations (Rubin B=20.6, Rubin R=1.05), and an 89.6% reduction in bias was achieved, with a remaining mean bias of 3.2%. The effect of chemoprophylaxis on EDH was insignificant ( P =0.294). Symptomatic EDH was independently associated with having a transfusion [odds ratio (OR)=7.30 (1.15, 46.20), P =0.035], having thoracic-level surgery [OR=41.19 (3.75, 452.4), P =0.002], and increasing body mass index [OR=1.44 (1.04, 1.98), P =0.028] but was not associated with chemoprophylaxis. Five out of 13 patients who developed EDH (38.5%) were receiving some form of anticoagulation, including 1 patient on therapeutic anticoagulation, 1 concurrently on aspirin and chemoprophylaxis, and 2 who were also found to have developed thrombocytopenia postoperatively. The median time on anticoagulation before EDH was 8.1 days. A higher proportion of patients who developed EDH also developed venous thromboembolic events than the general population [38.5% vs. 2.4%, OR=25.34 (9.226, 79.68), P <0.0001], and 1 EDH patient died from pulmonary embolism while off chemoprophylaxis. CONCLUSIONS: Chemoprophylactic anticoagulation did not cause an increase in the rate of spinal EDH in our patient population.

The demographic, clinical, and management differences between traumatic dens fracture patients with and without simultaneous atlas fractures.

Introduction: Atlas fractures often accompany traumatic dens fractures, but existing literature on the management of simultaneous atlantoaxial fractures is limited. Methods: We examined all patients with traumatic dens fractures at our institution between 2008 and 2018. We used multivariable logistic regression and ordinal logistic regression to identify factors independently associated with presentation with a simultaneous atlas fracture, as well myelopathy severity, fracture nonunion, and selection for surgery. Results: Two hundred and eighty-two patients with traumatic dens fractures without subaxial fractures were identified, including 65 (22.8%) with simultaneous atlas fractures. The distribution of injury mechanisms differed between groups (χ2 P = 0.0360). On multivariable logistic regression, dens nonunion was positively associated with type II fractures (odds ratio [OR] = 2.00, P = 0.038) and negatively associated with having surgery (OR = 0.52, P = 0.049), but not with having a C1 fracture (P = 0.3673). Worse myelopathy severity on presentation was associated with having a severe injury severity score (OR = 102.3, P < 0.001) and older age (OR = 1.28, P = 0.002), but not with having an atlas fracture (P = 0.2446). Having a simultaneous atlas fracture was associated with older age (OR = 1.29, P = 0.024) and dens fracture angulation (OR = 2.62, P = 0.004). Among patients who underwent surgery, C1/C2 posterior fusion was the most common procedure, and having a simultaneous atlas fracture was associated with selection for occipitocervical fusion (OCF) (OR = 14.35, P = 0.010). Conclusions: Among patients with traumatic dens, patients who have simultaneous atlas fractures are a distinct subpopulation with respect to age, mechanism of injury, fracture morphology, and management. Traumatic dens fractures with simultaneous atlas fractures are independently associated with selection for OCF rather than posterior cervical fusion alone.

Chemoprophylactic Anticoagulation Following Lumbar Surgery Significantly Reduces Thromboembolic Events After Instrumented Fusions, Not Decompressions.

STUDY DESIGN: Retrospective cohort. OBJECTIVE: To quantify any reduction in venous thromboembolic events (VTEs) caused by chemoprophylaxis among lumbar surgery patients. SUMMARY OF BACKGROUND DATA: Chemoprophylactic anticoagulation (chemoprophylaxis) is used to prevent VTE after lumbar surgery. However, the treatment effect of chemoprophylaxis has not been reported among spine surgery patients, as conventional statistical methods preclude such inferences. MATERIALS AND METHODS: A total of 1243 consecutive lumbar fusions and 1433 noninstrumented lumbar decompressions performed at our institution over a six-year period were identified, and clinical and demographic data were collected, including on VTE events within 30 days postoperatively. Instrumented lumbar fusions and noninstrumented lumbar surgeries were analyzed separately. Patients who were given chemoprophylaxis (treatment) and controls were matched according to known VTE risk factors, including age, body mass index, sex, diabetes, chronic kidney disease, history of VTE, estimated blood loss, length of surgery, transfusion, whether surgery was staged, and whether surgery used an anterior approach. K-nearest neighbor propensity score matching was performed, and the treatment effect of chemoprophylaxis was calculated. RESULTS: Unadjusted, there was no difference in the rate of VTE between treatment and controls in either population. Baseline clinical and demographic characteristics differed significantly between treatment and control groups. In all, 575 lumbar fusion patients and 435 noninstrumented lumbar decompression patients were successfully propensity score matched, yielding balanced models (Rubin B <25, 0.560% reduction in known bias for both populations. The treatment effect of chemoprophylaxis after lumbar fusion in our patient population was a reduction in VTE incidence from 9.4% to 4.2% ( P <0.05), and propensity score adjusted regression confirmed a reduced odds of VTE with chemoprophylaxis (odds ratio=0.37, P =0.035). The treatment effect was not significant for noninstrumented lumbar decompression patients. CONCLUSION: Among patients undergoing instrumented lumbar fusions, chemoprophylactic anticoagulation causes a significant reduction in VTE, but causes no significant reduction among patients undergoing noninstrumented lumbar decompression.

Variation in academic neurosurgery departments' #neurosurgery social media influence.

Background: Social media use is increasingly common among academic neurosurgery departments, but its relationship with academic metrics remains underexamined. Methods: We examine the relationship between American academic neurosurgery departments' number of followers on Twitter, Instagram, and Facebook and the following academic metrics: Doximity Residency rankings, US News & World Report rankings (USNWR) of their affiliated medical schools, and the amount of NIH funding of those schools. Results: Few departments had disproportionate number of followers. A greater proportion of programs had Twitter accounts (88.9%) than had Instagram (72.2%) or Facebook (51.9%) accounts (p=0.0001). Programs identified as "Influencers" had more departmental NIH funding (p=0.044), more institutional NIH funding (p=0.035), better Doximity residency rankings (p=0.044), and better affiliated medical school rankings (p=0.002). Number of Twitter followers had the strongest correlation with academic metrics, yet only modest correlations were identified to departmental NIH funding (R=0.496, p=0.0001), institutional NIH funding (R=0.387, p=0.0072), Doximity residency rank (R=0.411, p=0.0020), and affiliated medical school ranking (R=0.545,p<0.0001). On multivariable regression, only being affiliated with a medical school in the top quartile on the USNWR rankings, rather than neurosurgery departmental metrics, predicted having more Twitter (OR=5.666, p=0.012) and Instagram (OR=8.33, p=0.009) followers. Conclusion: American academic neurosurgery departments preferentially use Twitter over Instagram or Facebook. Their Twitter or Instagram presences are associated with better performance on traditional academic metrics. However, these associations are modest, suggesting that other factors contribute to a department's social media influence. A department's affiliated medical school may contribute to the department's social media brand.

Atlas fractures with and without simultaneous dens fractures differ with respect to clinical, demographic, and management characteristics.

Background: Patients with simultaneous fractures of the atlas and dens have traditionally been managed according to the dens fracture's morphology, but data supporting this practice are limited. Methods: We retrospectively examined all patients with traumatic atlas fractures at our institution between 2008 and 2016. We used multivariable regression and propensity score matching to compare the presentation, management, and outcomes of patients with isolated atlas fractures to patients with simultaneous atlas-dens fractures. Results: Ninety-nine patients were identified. Patients with isolated atlas fractures were younger (61 ± 22 vs. 77 ± 14, P = 0.0003), had lower median Charlson Comorbidity Index (3 vs. 5, P = 0.0005), had better presenting Nurick myelopathy scores (0 vs. 3, P < 0.0001), and had different mechanisms of injury (P = 0.0011). Multivariable regression showed that having a simultaneous atlas-dens fracture was independently associated with older age (odds ratio [OR] =1.59 [1.22, 2.07], P = 0.001), worse presenting myelopathy (OR = 3.10 [2.04, 4.16], P < 0.001), and selection for surgery (OR = 4.91 [1.10, 21.97], P = 0.037). Propensity score matching yielded balanced populations (Rubin's B = 23.3, Rubin's R = 1.96) and showed that the risk of atlas fracture nonunion was no different among isolated atlas fractures compared to simultaneous atlas-dens fractures (P = 0.304). Age was the only variable independently associated with atlas fracture nonunion (OR = 2.39 [1.15, 5.00], P = 0.020), having a simultaneous atlas-dens fracture was not significant (P = 0.2829). Conclusions: Among patients with atlas fractures, simultaneous fractures of the dens occur in older patients and confer an increased risk of myelopathy and requiring surgical stabilization. Controlling for confounders, the risk of atlas fracture nonunion is equivalent for isolated atlas fractures versus simultaneous atlas-dens fractures.

Outcome Reporting in Spine Surgery: A Review of Historical and Emerging Trends.

The general objectives of spine surgery are to alleviate pain, restore neurologic function, and prevent or treat spinal deformities or instability. The accumulating expanse of outcome measures has allowed us to more objectively quantify these variables and, therefore, gauge the success of treatments, ultimately improving the quality of the delivered health care. It has become increasingly evident that spinal conditions and their accompanying interventions affect all aspects of a patient's life, including their physical, mental, emotional, and social well-being. This underscores the challenge of creating clinically relevant and accurate outcome measures in spine care, and the reason why there is a growing recognition of the importance of subjective measures such as patient-reported outcome measures, that consider a patients' health-related quality of life. Subjective measures provide valuable insights into patient experiences and perceptions of treatment outcomes, whereas objective measures provide a reproducible glimpse into key radiographic and clinical parameters that are associated with a successful outcome. In this narrative review, we provide a detailed analysis of the most common subjective and objective outcome measures employed in spine surgery, with a special focus on their current role as well as the possible future of outcome reporting.

Traumatic Dens Fracture Patients Comprise Distinct Subpopulations Distinguished by Differences in Age, Sex, Injury Mechanism and Severity, and Outcome.

BACKGROUND: Dens fractures are an increasingly common injury, yet their epidemiology and its implications remain underexamined. METHODS: We retrospectively analyzed all traumatic dens fracture patients managed at our institution over a 10-year period, examining demographic, clinical, and outcomes data. Patient subsets were compared across these parameters. RESULTS: Among 303 traumatic dens fracture patients, we observed a bimodal age distribution with a strong goodness of fit centered at age 22.3 ± 5.7 (R = 0.8781) and at 77.7 ± 13.9 (R = 0.9686). A population pyramid demonstrated a bimodal distribution among male patients, but not female patients, which was confirmed with a strong goodness of fit for male patient subpopulations age <35 (R = 0.9791) and age ≥35 (R = 0.8843), but a weaker fit for a second female subpopulation age <35. Both age groups were equally likely to undergo surgery. Patients younger than age 35 were more likely to be male (82.4% vs. 46.9%, odds ratio [OR] = 5.29 [1.54, 17.57], P = 0.0052), have motor vehicle collision as their mechanism of injury (64.7% vs. 14.1%, OR = 11.18 [3.77, 31.77], P < 0.0001), and to have a severe trauma injury severity score (17.6% vs. 2.9%, OR = 7.23 [1.88, 28.88], P = 0.0198). Nevertheless, patients age <35 were less likely to have fracture nonunion at follow (18.2% vs. 53.7%, OR = 0.19 [0.041, 0.76], P = 0.0288). CONCLUSIONS: The dens fracture patient population comprises 2 subpopulations, distinguished by differences in age, sex, injury mechanism and severity, and outcome, with male dens fracture patients demonstrating a bimodal age distribution. Young, male patients were more likely to have high-energy injury mechanisms leading to severe trauma, yet were less likely to have fracture nonunion at follow-up.

Odontoid fracture type and angulation affect nonunion risk, but comminution and displacement do not: A propensity score matched analysis of fracture morphology.

OBJECTIVE: Odontoid fractures disproportionately affect older patients who have high surgical risk, but also high rates of fracture nonunion. To guide surgical decision-making, we quantified the effect of fracture morphology on nonunion among nonoperatively managed, traumatic, isolated odontoid fractures. METHODS: We examined all patients with isolated odontoid fractures treated nonoperatively at our institution between 2010 and 2019. Multivariable regression and propensity score matching were used to quantify the effect of fracture type, angulation, comminution, and displacement on bony healing by 26 weeks from injury. RESULTS: 303 consecutive traumatic odontoid fracture patients were identified, of whom 163 (53.8 %) had isolated fractures that were managed nonoperatively. Selection for nonoperative management was more likely with older age (OR=1.31 [1.09, 1.58], p = 0.004), and less likely with higher fracture angle (OR=0.70 [0.55, 0.89], p = 0.004), or higher presenting Nurick scores (OR=0.77 [0.62, 0.94], p = 0.011). Factors associated with nonunion at 26 weeks were fracture angle (OR=5.11 [1.43, 18.26], p = 0.012) and Anderson-D'Alonzo Type II morphology (OR=5.79 [1.88, 17.83], p = 0.002). Propensity score matching to assess the effect of type II fracture, fracture angulation> 10o, displacement≥ 3 mm, and comminution all yielded balanced models (Rubin's B<25.0, 0.5  10o (p = 0.015), and there was an 18.2 % lower rate of bony healing for each 10o increase in fracture angle. Fracture displacement≥ 3 mm and comminution had no significant effect. CONCLUSION: Type II fracture morphology and fracture angle > 10o significantly increase nonunion among nonoperatively managed isolated traumatic odontoid fractures, but fracture comminution and displacement ≥ 3 mm do not.

Surgery Decreases Nonunion, Myelopathy, and Mortality for Patients With Traumatic Odontoid Fractures: A Propensity Score Matched Analysis.

BACKGROUND: Existing literature suggests that surgical intervention for odontoid fractures is beneficial but often does not control for known confounding factors. OBJECTIVE: To examine the effect of surgical fixation on myelopathy, fracture nonunion, and mortality after traumatic odontoid fractures. METHODS: We analyzed all traumatic odontoid fractures managed at our institution between 2010 and 2020. Ordinal multivariable logistic regression was used to identify factors associated with myelopathy severity at follow-up. Propensity score analysis was used to test the treatment effect of surgery on nonunion and mortality. RESULTS: Three hundred and three patients with traumatic odontoid fracture were identified, of whom 21.6% underwent surgical stabilization. After propensity score matching, populations were well balanced across all analyses (Rubin's B < 25.0, 0.5 < Rubin's R < 2.0). Controlling for age and fracture angulation, type, comminution, and displacement, the overall rate of nonunion was lower in the surgical group (39.7% vs 57.3%, average treatment effect [ATE] = -0.153 [-0.279, -0.028], P = .017). Controlling for age, sex, Nurick score, Charlson Comorbidity Index, Injury Severity Score, and selection for intensive care unit admission, the mortality rate was lower for the surgical group at 30 days (1.7% vs 13.8%, ATE = -0.101 [-0.172, -0.030], P = .005) and at 1 year was 7.0% vs 23.7%, ATE = -0.099 [-0.181, -0.017], P = .018. Cox proportional hazards analysis also demonstrated a mortality benefit for surgery (hazard ratio = 0.587 [0.426, 0.799], P = .0009). Patients who underwent surgery were less likely to have worse myelopathy scores at follow-up (odds ratio = 0.48 [0.25, 0.93], P = .029). CONCLUSION: Surgical stabilization is associated with better myelopathy scores at follow-up and causes lower rates of fracture nonunion, 30-day mortality, and 1-year mortality.

Functional Anatomy and Biomechanics of the Craniovertebral Junction.

The craniovertebral junction (CVJ) involves the atlas, axis, and occiput along with the atlanto-occipital and atlantoaxial joints. The anatomy and neural and vascular anatomy of the junction render the CVJ unique. Specialists treating disorders that affect the CVJ must appreciate its intricate anatomy and should be well versed in its biomechanics. This first article in a three-article series provides an overview of the functional anatomy and biomechanics of the CVJ.

Adolescent Idiopathic Scoliosis and Pregnancy.

Objective This study examined the interaction between adolescent idiopathic scoliosis (AIS) and pregnancy, focusing on pregnancy outcomes, changes in back pain, and anesthesia use. Methods A retrospective analysis was conducted on adult patients with AIS who gave birth at our institution between 2006 and 2022. Results A total of 163 AIS patients with 263 pregnancies were included. The median age at delivery was 33 (range 18 to 50) years. Among 157 patients with information on prior scoliosis treatment, 66.9% had not received treatment, 20.4% had undergone spinal fusion, and 12.7% had received bracing. Of the 260 pregnancies with available data, 90.4% were delivered at term and 8.5% were preterm. Of the 257 pregnancies with information on anesthesia type, 35.0% received epidural anesthesia, 17.9% received spinal anesthesia, 37.7% received combined spinal and epidural anesthesia, 8.2% received no anesthesia, and 1.2% received intravenous or general anesthesia. Difficulty administering neuraxial anesthesia was reported in 6.1% of cases, and these patients were less likely to receive combined spinal and epidural anesthesia (6.3% versus 39.8%, p = 0.0123). Among 116 cases with recorded back pain during pregnancy, 67.2% reported increased pain, 31.9% reported similar pain, and one patient reported decreased pain. Of the 16 patients with pre and postpartum radiographs, eight showed a Cobb angle increase ≥ 3°, with five patients having an increase ≥ 5°. Conclusions Pregnancy can exacerbate back pain and pose challenges for neuraxial anesthesia in some AIS patients. Further large-scale, multi-institutional studies with standardized data collection are needed to fully understand the impact of pregnancy on AIS.

Advances, Challenges, and Future Directions in the Management of Craniovertebral Junction Pathologies.

In this third article in a 3-article series on the craniocervical junction, we define the terms "basilar impression," "cranial settling," "basilar invagination," and "platybasia," noting that these terms are often used interchangeably but represent distinct entities. We then provide examples that represent these pathologies and treatment paradigms. Finally, we discuss the challenges and future direction in the craniovertebral junction surgery space.

An Approach to Managing Disorders Affecting the Craniovertebral Junction.

The craniovertebral junction (CVJ), or the "first junction," can be affected by a variety of pathological states. Some of these conditions could represent a gray area in that they can be treated by general neurosurgeons or such specialists as skull base or spinal surgeons. However, some conditions are best managed with a multidisciplinary approach. The importance of in-depth knowledge of the anatomy and biomechanics of this junction cannot be overemphasized. Identifying what represents clinical stability or instability is key to successful diagnosis and, hence, treatment. In this report, the second in a 3-article series, we describe our approach to managing CVJ pathologies in a case-based fashion to illustrate key concepts.

Outpatient telemedicine in neurosurgery: 15,677 consecutive encounters in a comparative analysis of its effectiveness and impact on the surgical conversion rate.

OBJECTIVE: The coronavirus disease 2019 (COVID-19) pandemic has necessitated the use of telehealth visits (THVs). The effects on neurosurgical practice have not been well characterized, especially concerning new-patient THVs. Therefore, the authors of this study reviewed their institution's experience with outpatient clinic visits and THVs from before the COVID-19 pandemic to the present to focus on clinical metrics, rates of surgery, and the effects of implementing THVs in order to better understand their implications for clinical practice as more data emerge over time. METHODS: The authors reviewed 15,677 consecutive new outpatient in-person visits (IPVs), THVs, and neurosurgical procedures/cases proceeding from their institution between 2018 and 2022 for trends and associations related to THVs. RESULTS: Among spine patients, there was no difference in the proportion of encounters that led to surgery (surgical conversion rate) between THVs and IPVs (p = 0.49). Among cranial patients, THVs were negatively associated with conversion (OR 0.73, p = 0.03). On average, patients using THVs lived further from the hospital (p < 0.001); however, the patient catchment area appeared unchanged. The median distance to the hospital among THV patients was counterbalanced by a decreased distance for spine patients pursing IPVs (p < 0.001), with no significant change to case volume. There was no change in distance to the hospital among cranial patients. For both cranial and spine patients, surgical conversion was more likely among those who lived a great distance from the hospital if their initial encounter was an IPV (p = 0.007 and < 0.001, respectively). However, there was no relationship between distance from the hospital and surgical conversion among THV patients (p = 0.565). The availability of THVs did not significantly affect follow-up time (p = 0.837). For new patients at IPVs, there was no difference in time to the operating room between cranial and spine cases; for new patients at THVs, however, time to the operating room was significantly faster for cranial cases than for spine cases (p = 0.0018). CONCLUSIONS: Compared to IPVs, THVs lead to decreased surgical conversion for cranial patients but not spine patients. THVs do not appear to increase the catchment area. For patients who live far from the hospital, an IPV is associated with surgical conversion. Surgical conversion is faster following cranial THVs than after spine THVs. THVs did not increase the duration of follow-up.

Using machine learning and big data for the prediction of venous thromboembolic events after spine surgery: A single-center retrospective analysis of multiple models on a cohort of 6869 patients.

Objective: Venous thromboembolic event (VTE) after spine surgery is a rare but potentially devastating complication. With the advent of machine learning, an opportunity exists for more accurate prediction of such events to aid in prevention and treatment. Methods: Seven models were screened using 108 database variables and 62 preoperative variables. These models included deep neural network (DNN), DNN with synthetic minority oversampling technique (SMOTE), logistic regression, ridge regression, lasso regression, simple linear regression, and gradient boosting classifier. Relevant metrics were compared between each model. The top four models were selected based on area under the receiver operator curve; these models included DNN with SMOTE, linear regression, lasso regression, and ridge regression. Separate random sampling of each model was performed 1000 additional independent times using a randomly generated training/testing distribution. Variable weights and magnitudes were analyzed after sampling. Results: Using all patient-related variables, DNN using SMOTE was the top-performing model in predicting postoperative VTE after spinal surgery (area under the curve [AUC] =0.904), followed by lasso regression (AUC = 0.894), ridge regression (AUC = 0.873), and linear regression (AUC = 0.864). When analyzing a subset of only preoperative variables, the top-performing models were lasso regression (AUC = 0.865) and DNN with SMOTE (AUC = 0.864), both of which outperform any currently published models. Main model contributions relied heavily on variables associated with history of thromboembolic events, length of surgical/anesthetic time, and use of postoperative chemoprophylaxis. Conclusions: The current study provides promise toward machine learning methods geared toward predicting postoperative complications after spine surgery. Further study is needed in order to best quantify and model real-world risk for such events.

Traumatic atlas fracture patients comprise two subpopulations with distinct demographics and mechanisms of injury.

BACKGROUND: Atlas fractures account for as much as 13% of cervical fractures, yet their epidemiology and its implications remain under-examined. METHODS: We retrospectively analyzed 97 consecutive cases of isolated, traumatic atlas fractures at our institution over a 17-year period with respect to demographic, clinical, and outcomes data. Unique patient subsets were identified and compared across these parameters. RESULTS: The age of atlas fracture patients showed a bimodal distribution and strong goodness of fit, with one mean centered at an age of 30 years for patients age< 50 (R=0.9409) and mean age of 74 among patients age≥ 50 (R=0.8584). Young patients were more likely to have a high-risk mechanism of injury (57.8% vs. 11.5%, OR=10.49 [3.59, 29.65], p < 0.0001) and injuries while intoxicated (13.3% vs. 0%, OR ∞ [1.704, ∞], p = 0.0082). A greater portion of young patients were managed with halo (33.3% vs. 13.5%, OR=3.21 [1.20, 8.13, p = 0.0281]). Among patients who were managed with halo, a greater proportion had halo-related complications among patients age≥ 50 (57.1% vs. 6.7%, OR=18.67 [1.55, 239.1], p = 0.0207). The median age of atlas fractures increased by ~2.6 years annually (slope 2.637, p < 0.0001, R=0.8079). CONCLUSIONS: The atlas fracture patient population may comprise two distinct subpopulations, distinguished by differences in age and mechanism of injury that lead to divergent management decisions. While halo immobilization has a low rate of complications among patients age< 50, the complication among patients age≥ 50 was significantly higher. The median age of atlas fracture patients increased linearly during the study period, highlighting the importance of age-related differences in management.

Risk Factors for Fracture Nonunion and Transverse Atlantal Ligament Injury After Isolated Atlas Fractures: A Case Series of 97 Patients.

BACKGROUND: The management of atlas fractures is controversial and hinges on the integrity of transverse atlantal ligament (TAL). OBJECTIVE: To identify risk factors for atlas fracture nonunion, with and without TAL injury. METHODS: All isolated, traumatic atlas fractures treated at our institution between 1999 and 2016 were analyzed. Multivariable logistic regression was used to identify variables associated with TAL injury confirmed on MRI, occult TAL injury seen on MRI but not suspected on computed tomography (CT), and with fracture nonunion on follow-up CT at 12 weeks. RESULTS: Lateral mass displacement (LMD) ≥ 7 mm had a 48.2% sensitivity, 98.3% specificity, and 82.6% accuracy for identifying TAL injury. MRI-confirmed TAL injury was independently associated with LMD > 7 mm ( P = .004) and atlanto-dental interval ( P = .039), and occult TAL injury was associated with atlanto-dental interval ( P = .019). Halo immobilization was associated with having a Gehweiler type 3 fracture ( P = .020), a high-risk injury mechanism ( P = .023), and an 18.1% complication rate. Thirteen patients with TAL injury on MRI and/or LMD ≥ 7 mm were treated with a cervical collar only, and 11 patients (84.6%) healed at 12 weeks. Nonunion rates at 12 weeks were equivalent between halo (11.1%) and cervical collar (12.5%). Only age independently predicted nonunion at 12 weeks ( P = .026). CONCLUSION: LMD > 7 mm on CT is not sensitive for TAL injury. Some atlas fractures with TAL injury can be managed with a cervical collar. Nonunion rates are not different between halo immobilization and cervical collar, but a strong selection bias precludes directly comparing the efficacy of these modalities. Age independently predicts nonunion.