Member (Postdoc), Cardiovascular Institute
MD, Icahn School of Medicine at Mount Sinai (2021)
BS, Georgia Institute of Technology, Chemical and Biomolecular Engineering (2015)
Douglas B. Unis, Sulaiman Somani, Anthony B. Costa. "United States Patent 10,945,848 Apparatus, method and system for providing customizable bone implants", Icahn School of Medicine at Mount Sinai, Mar 16, 2021
Current Research and Scholarly Interests
computer vision, personalized medicine, cardiology, clinical informatics
Enhancing convolutional neural network predictions of electrocardiograms with left ventricular dysfunction using a novel sub-waveform representation.
Cardiovascular digital health journal
2022; 3 (5): 220-231
Background: Electrocardiogram (ECG) deep learning (DL) has promise to improve the outcomes of patients with cardiovascular abnormalities. In ECG DL, researchers often use convolutional neural networks (CNNs) and traditionally use the full duration of raw ECG waveforms that create redundancies in feature learning and result in inaccurate predictions with large uncertainties.Objective: For enhancing these predictions, we introduced a sub-waveform representation that leverages the rhythmic pattern of ECG waveforms (data-centric approach) rather than changing the CNN architecture (model-centric approach).Results: We applied the proposed representation to a population with 92,446 patients to identify left ventricular dysfunction. We found that the sub-waveform representation increases the performance metrics compared to the full-waveform representation. We observed a 2% increase for area under the receiver operating characteristic curve and 10% increase for area under the precision-recall curve. We also carefully examined three reliability components of explainability, interpretability, and fairness. We provided an explanation for enhancements obtained by heartbeat alignment mechanism. By developing a new scoring system, we interpreted the clinical relevance of ECG features and showed that sub-waveform representation further pushes the scores towards clinical predictions. Finally, we showed that the new representation significantly reduces prediction uncertainties within subgroups that contributes to individual fairness.Conclusion: We expect that this added control over the granularity of ECG data will improve the DL modeling for new artificial intelligence technologies in the cardiovascular space.
View details for DOI 10.1016/j.cvdhj.2022.07.074
View details for PubMedID 36310683
Population scale latent space cohort matching for the improved use and exploration of observational trial data.
Mathematical biosciences and engineering : MBE
2022; 19 (7): 6795-6813
A significant amount of clinical research is observational by nature and derived from medical records, clinical trials, and large-scale registries. While there is no substitute for randomized, controlled experimentation, such experiments or trials are often costly, time consuming, and even ethically or practically impossible to execute. Combining classical regression and structural equation modeling with matching techniques can leverage the value of observational data. Nevertheless, identifying variables of greatest interest in high-dimensional data is frequently challenging, even with application of classical dimensionality reduction and/or propensity scoring techniques. Here, we demonstrate that projecting high-dimensional medical data onto a lower-dimensional manifold using deep autoencoders and post-hoc generation of treatment/control cohorts based on proximity in the lower-dimensional space results in better matching of confounding variables compared to classical propensity score matching (PSM) in the original high-dimensional space (P<0.0001) and performs similarly to PSM models constructed by experts with prior knowledge of the underlying pathology when evaluated on predicting risk ratios from real-world clinical data. Thus, in cases when the underlying problem is poorly understood and the data is high-dimensional in nature, matching in the autoencoder latent space might be of particular benefit.
View details for DOI 10.3934/mbe.2022320
View details for PubMedID 35730283
Development of a machine learning model using electrocardiogram signals to improve acute pulmonary embolism screening.
European heart journal. Digital health
2022; 3 (1): 56-66
Aims: Clinical scoring systems for pulmonary embolism (PE) screening have low specificity and contribute to computed tomography pulmonary angiogram (CTPA) overuse. We assessed whether deep learning models using an existing and routinely collected data modality, electrocardiogram (ECG) waveforms, can increase specificity for PE detection.Methods and results: We create a retrospective cohort of 21 183 patients at moderate- to high suspicion of PE and associate 23 793 CTPAs (10.0% PE-positive) with 320 746 ECGs and encounter-level clinical data (demographics, comorbidities, vital signs, and labs). We develop three machine learning models to predict PE likelihood: an ECG model using only ECG waveform data, an EHR model using tabular clinical data, and a Fusion model integrating clinical data and an embedded representation of the ECG waveform. We find that a Fusion model [area under the receiver-operating characteristic curve (AUROC) 0.81±0.01] outperforms both the ECG model (AUROC 0.59±0.01) and EHR model (AUROC 0.65±0.01). On a sample of 100 patients from the test set, the Fusion model also achieves greater specificity (0.18) and performance (AUROC 0.84±0.01) than four commonly evaluated clinical scores: Wells' Criteria, Revised Geneva Score, Pulmonary Embolism Rule-Out Criteria, and 4-Level Pulmonary Embolism Clinical Probability Score (AUROC 0.50-0.58, specificity 0.00-0.05). The model is superior to these scores on feature sensitivity analyses (AUROC 0.66-0.84) and achieves comparable performance across sex (AUROC 0.81) and racial/ethnic (AUROC 0.77-0.84) subgroups.Conclusion: Synergistic deep learning of ECG waveforms with traditional clinical variables can increase the specificity of PE detection in patients at least at moderate suspicion for PE.
View details for DOI 10.1093/ehjdh/ztab101
View details for PubMedID 35355847
Population scale latent space cohort matching for the improved use and exploration of observational trial data
MATHEMATICAL BIOSCIENCES AND ENGINEERING
2022; 19 (7): 6795-6813
View details for DOI 10.3934/mbe.2022320
View details for Web of Science ID 000797466800002
Using Deep-Learning Algorithms to Simultaneously Identify Right and Left Ventricular Dysfunction From the Electrocardiogram.
JACC. Cardiovascular imaging
OBJECTIVES: This study sought to develop DL models capable of comprehensively quantifying left and right ventricular dysfunction from ECG data in a large, diverse population.BACKGROUND: Rapid evaluation of left and right ventricular function using deep learning (DL) on electrocardiograms (ECGs) can assist diagnostic workflow. However, DL tools to estimate right- ventricular (RV) function do not exist, whereas those to estimate left ventricular (LV) function are restricted to quantification of very low LV function only.METHODS: A multicenter study was conducted with data from 5 New York City hospitals: 4 for internal testing and 1serving as external validation. We created novel DL models to classify left ventricular ejection fraction (LVEF) into categories derived from the latest universal definition of heart failure, estimate LVEF through regression, and predict a composite outcome of either RV systolic dysfunction or RV dilation.RESULTS: We obtained echocardiogram LVEF estimates for 147,636 patients paired to 715,890 ECGs. We used natural language processing (NLP) to extract RV size and systolic function information from 404,502 echocardiogram reports paired to 761,510 ECGs for 148,227 patients. For LVEF classification in internal testing, area under curve (AUC) at detection of LVEF≤40%, 40%< LVEF≤50%, and LVEF >50% was 0.94 (95%CI: 0.94-0.94), 0.82 (95%CI: 0.81-0.83), and 0.89 (95%CI: 0.89-0.89), respectively. For external validation, these results were 0.94 (95%CI: 0.94-0.95), 0.73 (95%CI: 0.72-0.74), and 0.87 (95%CI: 0.87-0.88). For regression, the mean absolute error was 5.84% (95%CI: 5.82%-5.85%) for internal testing and 6.14% (95%CI: 6.13%-6.16%) in external validation. For prediction of the composite RV outcome, AUC was 0.84 (95%CI: 0.84-0.84) in both internal testing and external validation.CONCLUSIONS: DL on ECG data can be used to create inexpensive screening, diagnostic, and predictive tools for both LV and RV dysfunction. Such tools may bridge the applicability of ECGs and echocardiography and enable prioritization of patients for further interventions for either sided failure progressing to biventricular disease.
View details for DOI 10.1016/j.jcmg.2021.08.004
View details for PubMedID 34656465
Health Equity in Artificial Intelligence and Primary Care Research: Protocol for a Scoping Review.
JMIR research protocols
2021; 10 (9): e27799
BACKGROUND: Though artificial intelligence (AI) has the potential to augment the patient-physician relationship in primary care, bias in intelligent health care systems has the potential to differentially impact vulnerable patient populations.OBJECTIVE: The purpose of this scoping review is to summarize the extent to which AI systems in primary care examine the inherent bias toward or against vulnerable populations and appraise how these systems have mitigated the impact of such biases during their development.METHODS: We will conduct a search update from an existing scoping review to identify studies on AI and primary care in the following databases: Medline-OVID, Embase, CINAHL, Cochrane Library, Web of Science, Scopus, IEEE Xplore, ACM Digital Library, MathSciNet, AAAI, and arXiv. Two screeners will independently review all abstracts, titles, and full-text articles. The team will extract data using a structured data extraction form and synthesize the results in accordance with PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines.RESULTS: This review will provide an assessment of the current state of health care equity within AI for primary care. Specifically, we will identify the degree to which vulnerable patients have been included, assess how bias is interpreted and documented, and understand the extent to which harmful biases are addressed. As of October 2020, the scoping review is in the title- and abstract-screening stage. The results are expected to be submitted for publication in fall 2021.CONCLUSIONS: AI applications in primary care are becoming an increasingly common tool in health care delivery and in preventative care efforts for underserved populations. This scoping review would potentially show the extent to which studies on AI in primary care employ a health equity lens and take steps to mitigate bias.INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/27799.
View details for DOI 10.2196/27799
View details for PubMedID 34533458
Predictive Approaches for Acute Dialysis Requirement and Death in COVID-19.
Clinical journal of the American Society of Nephrology : CJASN
2021; 16 (8): 1158-1168
BACKGROUND AND OBJECTIVES: AKI treated with dialysis initiation is a common complication of coronavirus disease 2019 (COVID-19) among hospitalized patients. However, dialysis supplies and personnel are often limited.DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: Using data from adult patients hospitalized with COVID-19 from five hospitals from the Mount Sinai Health System who were admitted between March 10 and December 26, 2020, we developed and validated several models (logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme GradientBoosting [XGBoost; with and without imputation]) for predicting treatment with dialysis or death at various time horizons (1, 3, 5, and 7 days) after hospital admission. Patients admitted to the Mount Sinai Hospital were used for internal validation, whereas the other hospitals formed part of the external validation cohort. Features included demographics, comorbidities, and laboratory and vital signs within 12 hours of hospital admission.RESULTS: A total of 6093 patients (2442 in training and 3651 in external validation) were included in the final cohort. Of the different modeling approaches used, XGBoost without imputation had the highest area under the receiver operating characteristic (AUROC) curve on internal validation (range of 0.93-0.98) and area under the precision-recall curve (AUPRC; range of 0.78-0.82) for all time points. XGBoost without imputation also had the highest test parameters on external validation (AUROC range of 0.85-0.87, and AUPRC range of 0.27-0.54) across all time windows. XGBoost without imputation outperformed all models with higher precision and recall (mean difference in AUROC of 0.04; mean difference in AUPRC of 0.15). Features of creatinine, BUN, and red cell distribution width were major drivers of the model's prediction.CONCLUSIONS: An XGBoost model without imputation for prediction of a composite outcome of either death or dialysis in patients positive for COVID-19 had the best performance, as compared with standard and other machine learning models.PODCAST: This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2021_07_09_CJN17311120.mp3.
View details for DOI 10.2215/CJN.17311120
View details for PubMedID 34031183
Development and validation of techniques for phenotyping ST-elevation myocardial infarction encounters from electronic health records.
2021; 4 (3): ooab068
Objectives: Classifying hospital admissions into various acute myocardial infarction phenotypes in electronic health records (EHRs) is a challenging task with strong research implications that remains unsolved. To our knowledge, this study is the first study to design and validate phenotyping algorithms using cardiac catheterizations to identify not only patients with a ST-elevation myocardial infarction (STEMI), but the specific encounter when it occurred.Materials and Methods: We design and validate multi-modal algorithms to phenotype STEMI on a multicenter EHR containing 5.1 million patients and 115 million patient encounters by using discharge summaries, diagnosis codes, electrocardiography readings, and the presence of cardiac catheterizations on the encounter.Results: We demonstrate that robustly phenotyping STEMIs by selecting discharge summaries containing "STEM" has the potential to capture the most number of STEMIs (positive predictive value [PPV] = 0.36, N=2110), but that addition of a STEMI-related International Classification of Disease (ICD) code and cardiac catheterizations to these summaries yields the highest precision (PPV = 0.94, N=952).Discussion and Conclusion: In this study, we demonstrate that the incorporation of percutaneous coronary intervention increases the PPV for detecting STEMI-related patient encounters from the EHR.
View details for DOI 10.1093/jamiaopen/ooab068
View details for PubMedID 34423260
Relational Learning Improves Prediction of Mortality in COVID-19 in the Intensive Care Unit.
IEEE transactions on big data
2021; 7 (1): 38-44
Traditional Machine Learning (ML) models have had limited success in predicting Coronoavirus-19 (COVID-19) outcomes using Electronic Health Record (EHR) data partially due to not effectively capturing the inter-connectivity patterns between various data modalities. In this work, we propose a novel framework that utilizes relational learning based on a heterogeneous graph model (HGM) for predicting mortality at different time windows in COVID-19 patients within the intensive care unit (ICU). We utilize the EHRs of one of the largest and most diverse patient populations across five hospitals in major health system in New York City. In our model, we use an LSTM for processing time varying patient data and apply our proposed relational learning strategy in the final output layer along with other static features. Here, we replace the traditional softmax layer with a Skip-Gram relational learning strategy to compare the similarity between a patient and outcome embedding representation. We demonstrate that the construction of a HGM can robustly learn the patterns classifying patient representations of outcomes through leveraging patterns within the embeddings of similar patients. Our experimental results show that our relational learning-based HGM model achieves higher area under the receiver operating characteristic curve (auROC) than both comparator models in all prediction time windows, with dramatic improvements to recall.
View details for DOI 10.1109/tbdata.2020.3048644
View details for PubMedID 33768136
View details for PubMedCentralID PMC7990133
Contrastive Learning Improves Critical Event Prediction in COVID-19 Patients.
Patterns (New York, N.Y.)
Deep Learning (DL) models typically require large-scale, balanced training data to be robust, generalizable, and effective in the context of healthcare. This has been a major issue for developing DL models for the coronavirus-disease 2019 (COVID-19) pandemic where data are highly class imbalanced. Conventional approaches in DL use cross-entropy loss (CEL) which often suffers from poor margin classification. We show that contrastive loss (CL) improves the performance of CEL especially in imbalanced electronic health records (EHR) data for COVID-19 analyses. We use a diverse EHR data set to predict three outcomes: mortality, intubation, and intensive care unit (ICU) transfer in hospitalized COVID-19 patients over multiple time windows. To compare the performance of CEL and CL, models are tested on the full data set and a restricted data set. CL models consistently outperform CEL models with differences ranging from 0.04 to 0.15 for AUPRC and 0.05 to 0.1 for AUROC.
View details for DOI 10.1016/j.patter.2021.100389
View details for PubMedID 34723227
View details for PubMedCentralID PMC8542449
AKI in Hospitalized Patients with COVID-19.
Journal of the American Society of Nephrology : JASN
2021; 32 (1): 151-160
Early reports indicate that AKI is common among patients with coronavirus disease 2019 (COVID-19) and associated with worse outcomes. However, AKI among hospitalized patients with COVID-19 in the United States is not well described.This retrospective, observational study involved a review of data from electronic health records of patients aged ≥18 years with laboratory-confirmed COVID-19 admitted to the Mount Sinai Health System from February 27 to May 30, 2020. We describe the frequency of AKI and dialysis requirement, AKI recovery, and adjusted odds ratios (aORs) with mortality.Of 3993 hospitalized patients with COVID-19, AKI occurred in 1835 (46%) patients; 347 (19%) of the patients with AKI required dialysis. The proportions with stages 1, 2, or 3 AKI were 39%, 19%, and 42%, respectively. A total of 976 (24%) patients were admitted to intensive care, and 745 (76%) experienced AKI. Of the 435 patients with AKI and urine studies, 84% had proteinuria, 81% had hematuria, and 60% had leukocyturia. Independent predictors of severe AKI were CKD, men, and higher serum potassium at admission. In-hospital mortality was 50% among patients with AKI versus 8% among those without AKI (aOR, 9.2; 95% confidence interval, 7.5 to 11.3). Of survivors with AKI who were discharged, 35% had not recovered to baseline kidney function by the time of discharge. An additional 28 of 77 (36%) patients who had not recovered kidney function at discharge did so on posthospital follow-up.AKI is common among patients hospitalized with COVID-19 and is associated with high mortality. Of all patients with AKI, only 30% survived with recovery of kidney function by the time of discharge.
View details for DOI 10.1681/ASN.2020050615
View details for PubMedID 32883700
View details for PubMedCentralID PMC7894657
Outcomes of Patients on Maintenance Dialysis Hospitalized with COVID-19.
Clinical journal of the American Society of Nephrology : CJASN
2021; 16 (3): 452-455
View details for DOI 10.2215/CJN.12360720
View details for PubMedID 33127607
View details for PubMedCentralID PMC8011022
Blood Donation and COVID-19: Reconsidering the 3-Month Deferral Policy for Gay, Bisexual, Transgender, and Other Men Who Have Sex With Men.
American journal of public health
2021; 111 (2): 247-252
In April 2020, in light of COVID-19-related blood shortages, the US Food and Drug Administration (FDA) reduced the deferral period for men who have sex with men (MSM) from its previous duration of 1 year to 3 months.Although originally born out of necessity, the decades-old restrictions on MSM donors have been mitigated by significant advancements in HIV screening, treatment, and public education. The severity of the ongoing COVID-19 pandemic-and the urgent need for safe blood products to respond to such crises-demands an immediate reconsideration of the 3-month deferral policy for MSM.We review historical HIV testing and transmission evidence, discuss the ethical ramifications of the current deferral period, and examine the issue of noncompliance with donor deferral rules. We also propose an eligibility screening format that involves an individual risk-based screening protocol and, unlike current FDA guidelines, does not effectively exclude donors on the basis of gender identity or sexual orientation. Our policy proposal would allow historically marginalized community members to participate with dignity in the blood donation process without compromising blood donation and transfusion safety outcomes.
View details for DOI 10.2105/AJPH.2020.305974
View details for PubMedID 33211588
View details for PubMedCentralID PMC7811078
Deep learning and the electrocardiogram: review of the current state-of-the-art.
Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology
In the recent decade, deep learning, a subset of artificial intelligence and machine learning, has been used to identify patterns in big healthcare datasets for disease phenotyping, event predictions, and complex decision making. Public datasets for electrocardiograms (ECGs) have existed since the 1980s and have been used for very specific tasks in cardiology, such as arrhythmia, ischemia, and cardiomyopathy detection. Recently, private institutions have begun curating large ECG databases that are orders of magnitude larger than the public databases for ingestion by deep learning models. These efforts have demonstrated not only improved performance and generalizability in these aforementioned tasks but also application to novel clinical scenarios. This review focuses on orienting the clinician towards fundamental tenets of deep learning, state-of-the-art prior to its use for ECG analysis, and current applications of deep learning on ECGs, as well as their limitations and future areas of improvement.
View details for DOI 10.1093/europace/euaa377
View details for PubMedID 33564873
Federated Learning of Electronic Health Records to Improve Mortality Prediction in Hospitalized Patients With COVID-19: Machine Learning Approach.
JMIR medical informatics
2021; 9 (1): e24207
Machine learning models require large datasets that may be siloed across different health care institutions. Machine learning studies that focus on COVID-19 have been limited to single-hospital data, which limits model generalizability.We aimed to use federated learning, a machine learning technique that avoids locally aggregating raw clinical data across multiple institutions, to predict mortality in hospitalized patients with COVID-19 within 7 days.Patient data were collected from the electronic health records of 5 hospitals within the Mount Sinai Health System. Logistic regression with L1 regularization/least absolute shrinkage and selection operator (LASSO) and multilayer perceptron (MLP) models were trained by using local data at each site. We developed a pooled model with combined data from all 5 sites, and a federated model that only shared parameters with a central aggregator.The LASSOfederated model outperformed the LASSOlocal model at 3 hospitals, and the MLPfederated model performed better than the MLPlocal model at all 5 hospitals, as determined by the area under the receiver operating characteristic curve. The LASSOpooled model outperformed the LASSOfederated model at all hospitals, and the MLPfederated model outperformed the MLPpooled model at 2 hospitals.The federated learning of COVID-19 electronic health record data shows promise in developing robust predictive models without compromising patient privacy.
View details for DOI 10.2196/24207
View details for PubMedID 33400679
View details for PubMedCentralID PMC7842859
Retrospective cohort study of clinical characteristics of 2199 hospitalised patients with COVID-19 in New York City.
2020; 10 (11): e040736
The COVID-19 pandemic is a global public health crisis, with over 33 million cases and 999 000 deaths worldwide. Data are needed regarding the clinical course of hospitalised patients, particularly in the USA. We aimed to compare clinical characteristic of patients with COVID-19 who had in-hospital mortality with those who were discharged alive.Demographic, clinical and outcomes data for patients admitted to five Mount Sinai Health System hospitals with confirmed COVID-19 between 27 February and 2 April 2020 were identified through institutional electronic health records. We performed a retrospective comparative analysis of patients who had in-hospital mortality or were discharged alive.All patients were admitted to the Mount Sinai Health System, a large quaternary care urban hospital system.Participants over the age of 18 years were included.We investigated in-hospital mortality during the study period.A total of 2199 patients with COVID-19 were hospitalised during the study period. As of 2 April, 1121 (51%) patients remained hospitalised, and 1078 (49%) completed their hospital course. Of the latter, the overall mortality was 29%, and 36% required intensive care. The median age was 65 years overall and 75 years in those who died. Pre-existing conditions were present in 65% of those who died and 46% of those discharged. In those who died, the admission median lymphocyte percentage was 11.7%, D-dimer was 2.4 μg/mL, C reactive protein was 162 mg/L and procalcitonin was 0.44 ng/mL. In those discharged, the admission median lymphocyte percentage was 16.6%, D-dimer was 0.93 μg/mL, C reactive protein was 79 mg/L and procalcitonin was 0.09 ng/mL.In our cohort of hospitalised patients, requirement of intensive care and mortality were high. Patients who died typically had more pre-existing conditions and greater perturbations in inflammatory markers as compared with those who were discharged.
View details for DOI 10.1136/bmjopen-2020-040736
View details for PubMedID 33247020
View details for PubMedCentralID PMC7702220
Machine Learning to Predict Mortality and Critical Events in a Cohort of Patients With COVID-19 in New York City: Model Development and Validation.
Journal of medical Internet research
2020; 22 (11): e24018
COVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking.The aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points.We used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19-positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions.Upon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction.We externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes.
View details for DOI 10.2196/24018
View details for PubMedID 33027032
View details for PubMedCentralID PMC7652593
Federated Learning of Electronic Health Records Improves Mortality Prediction in Patients Hospitalized with COVID-19.
medRxiv : the preprint server for health sciences
Machine learning (ML) models require large datasets which may be siloed across different healthcare institutions. Using federated learning, a ML technique that avoids locally aggregating raw clinical data across multiple institutions, we predict mortality within seven days in hospitalized COVID-19 patients. Patient data was collected from Electronic Health Records (EHRs) from five hospitals within the Mount Sinai Health System (MSHS). Logistic Regression with L1 regularization (LASSO) and Multilayer Perceptron (MLP) models were trained using local data at each site, a pooled model with combined data from all five sites, and a federated model that only shared parameters with a central aggregator. Both the federated LASSO and federated MLP models performed better than their local model counterparts at four hospitals. The federated MLP model also outperformed the federated LASSO model at all hospitals. Federated learning shows promise in COVID-19 EHR data to develop robust predictive models without compromising patient privacy.
View details for DOI 10.1101/2020.08.11.20172809
View details for PubMedID 32817979
View details for PubMedCentralID PMC7430624
Characterization of Patients Who Return to Hospital Following Discharge from Hospitalization for COVID-19.
Journal of general internal medicine
2020; 35 (10): 2838-2844
Data on patients with coronavirus disease 2019 (COVID-19) who return to hospital after discharge are scarce. Characterization of these patients may inform post-hospitalization care.To describe clinical characteristics of patients with COVID-19 who returned to the emergency department (ED) or required readmission within 14 days of discharge.Retrospective cohort study of SARS-COV-2-positive patients with index hospitalization between February 27 and April 12, 2020, with ≥ 14-day follow-up. Significance was defined as P < 0.05 after multiplying P by 125 study-wide comparisons.Hospitalized patients with confirmed SARS-CoV-2 discharged alive from five New York City hospitals.Readmission or return to ED following discharge.Of 2864 discharged patients, 103 (3.6%) returned for emergency care after a median of 4.5 days, with 56 requiring inpatient readmission. The most common reason for return was respiratory distress (50%). Compared with patients who did not return, there were higher proportions of COPD (6.8% vs 2.9%) and hypertension (36% vs 22.1%) among those who returned. Patients who returned also had a shorter median length of stay (LOS) during index hospitalization (4.5 [2.9,9.1] vs 6.7 [3.5, 11.5] days; Padjusted = 0.006), and were less likely to have required intensive care on index hospitalization (5.8% vs 19%; Padjusted = 0.001). A trend towards association between absence of in-hospital treatment-dose anticoagulation on index admission and return to hospital was also observed (20.9% vs 30.9%, Padjusted = 0.06). On readmission, rates of intensive care and death were 5.8% and 3.6%, respectively.Return to hospital after admission for COVID-19 was infrequent within 14 days of discharge. The most common cause for return was respiratory distress. Patients who returned more likely had COPD and hypertension, shorter LOS on index-hospitalization, and lower rates of in-hospital treatment-dose anticoagulation. Future studies should focus on whether these comorbid conditions, longer LOS, and anticoagulation are associated with reduced readmissions.
View details for DOI 10.1007/s11606-020-06120-6
View details for PubMedID 32815060
View details for PubMedCentralID PMC7437962
Coronavirus 2019 and People Living With Human Immunodeficiency Virus: Outcomes for Hospitalized Patients in New York City.
Clinical infectious diseases : an official publication of the Infectious Diseases Society of America
2020; 71 (11): 2933-2938
There are limited data regarding the clinical impact of coronavirus disease 2019 (COVID-19) on people living with human immunodeficiency virus (PLWH). In this study, we compared outcomes for PLWH with COVID-19 to a matched comparison group.We identified 88 PLWH hospitalized with laboratory-confirmed COVID-19 in our hospital system in New York City between 12 March and 23 April 2020. We collected data on baseline clinical characteristics, laboratory values, HIV status, treatment, and outcomes from this group and matched comparators (1 PLWH to up to 5 patients by age, sex, race/ethnicity, and calendar week of infection). We compared clinical characteristics and outcomes (death, mechanical ventilation, hospital discharge) for these groups, as well as cumulative incidence of death by HIV status.Patients did not differ significantly by HIV status by age, sex, or race/ethnicity due to the matching algorithm. PLWH hospitalized with COVID-19 had high proportions of HIV virologic control on antiretroviral therapy. PLWH had greater proportions of smoking (P < .001) and comorbid illness than uninfected comparators. There was no difference in COVID-19 severity on admission by HIV status (P = .15). Poor outcomes for hospitalized PLWH were frequent but similar to proportions in comparators; 18% required mechanical ventilation and 21% died during follow-up (compared with 23% and 20%, respectively). There was similar cumulative incidence of death over time by HIV status (P = .94).We found no differences in adverse outcomes associated with HIV infection for hospitalized COVID-19 patients compared with a demographically similar patient group.
View details for DOI 10.1093/cid/ciaa880
View details for PubMedID 32594164
View details for PubMedCentralID PMC7337691
Prevalence and Impact of Myocardial Injury in Patients Hospitalized With COVID-19 Infection.
Journal of the American College of Cardiology
2020; 76 (5): 533-546
The degree of myocardial injury, as reflected by troponin elevation, and associated outcomes among U.S. hospitalized patients with coronavirus disease-2019 (COVID-19) are unknown.The purpose of this study was to describe the degree of myocardial injury and associated outcomes in a large hospitalized cohort with laboratory-confirmed COVID-19.Patients with COVID-19 admitted to 1 of 5 Mount Sinai Health System hospitals in New York City between February 27, 2020, and April 12, 2020, with troponin-I (normal value <0.03 ng/ml) measured within 24 h of admission were included (n = 2,736). Demographics, medical histories, admission laboratory results, and outcomes were captured from the hospitals' electronic health records.The median age was 66.4 years, with 59.6% men. Cardiovascular disease (CVD), including coronary artery disease, atrial fibrillation, and heart failure, was more prevalent in patients with higher troponin concentrations, as were hypertension and diabetes. A total of 506 (18.5%) patients died during hospitalization. In all, 985 (36%) patients had elevated troponin concentrations. After adjusting for disease severity and relevant clinical factors, even small amounts of myocardial injury (e.g., troponin I >0.03 to 0.09 ng/ml; n = 455; 16.6%) were significantly associated with death (adjusted hazard ratio: 1.75; 95% CI: 1.37 to 2.24; p < 0.001) while greater amounts (e.g., troponin I >0.09 ng/dl; n = 530; 19.4%) were significantly associated with higher risk (adjusted HR: 3.03; 95% CI: 2.42 to 3.80; p < 0.001).Myocardial injury is prevalent among patients hospitalized with COVID-19; however, troponin concentrations were generally present at low levels. Patients with CVD are more likely to have myocardial injury than patients without CVD. Troponin elevation among patients hospitalized with COVID-19 is associated with higher risk of mortality.
View details for DOI 10.1016/j.jacc.2020.06.007
View details for PubMedID 32517963
View details for PubMedCentralID PMC7279721
Machine Learning in Cardiology-Ensuring Clinical Impact Lives Up to the Hype.
Journal of cardiovascular pharmacology and therapeutics
2020; 25 (5): 379-390
Despite substantial advances in the study, treatment, and prevention of cardiovascular disease, numerous challenges relating to optimally screening, diagnosing, and managing patients remain. Simultaneous improvements in computing power, data storage, and data analytics have led to the development of new techniques to address these challenges. One powerful tool to this end is machine learning (ML), which aims to algorithmically identify and represent structure within data. Machine learning's ability to efficiently analyze large and highly complex data sets make it a desirable investigative approach in modern biomedical research. Despite this potential and enormous public and private sector investment, few prospective studies have demonstrated improved clinical outcomes from this technology. This is particularly true in cardiology, despite its emphasis on objective, data-driven results. This threatens to stifle ML's growth and use in mainstream medicine. We outline the current state of ML in cardiology and outline methods through which impactful and sustainable ML research can occur. Following these steps can ensure ML reaches its potential as a transformative technology in medicine.
View details for DOI 10.1177/1074248420928651
View details for PubMedID 32495652
Preoperative Nutritional Status as a Risk Factor for Major Postoperative Complications Following Anterior Lumbar Interbody Fusion.
Global spine journal
2018; 8 (7): 662-667
Retrospective study.To determine rates of medical and surgical postoperative complications in adults with hypoalbuminemia undergoing anterior lumbar interbody fusion (ALIF).This was a retrospective analysis of prospectively collected data from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database of patients (≥18 years old) undergoing ALIF procedures, identified by CPT (Current Procedural Terminology) code from 2011 to 2014. Poor nutritional status was defined by a preoperative serum albumin level <3.5 g/dL, and albumin levels above this were considered normal. Multivariate logistic regression models were utilized to assess preoperative risk factors including nutritional status as predictors of specific postoperative complications. Significance was defined as P < .05 and odds ratios (ORs) were calculated with a 95% confidence interval (CI). This model was used to determine the strength of nutritional status as an adjusted predictor of adverse postoperative events.There were 3184 ALIF cases, including 1,275 (40%) of which had preoperative serum albumin levels. 53 (4.15%) patients were classified as having poor nutrition status. Poor preoperative nutritional status was shown to be a strong independent predictor of length of stay ≥5 days (OR = 2.56, 95% CI 1.43-4.59, P = .002), urinary tract infection (OR = 5.93, 95% CI 2.11-16.68, P = .001), and sepsis (OR = 5.35, 95% CI 1.13-25.42, P = .035) compared to patients with normal preoperative serum albumin levels.Our analysis shows that patients with poor nutritional status before ALIF are independently at risk for sepsis as well as increased length of stay and urinary tract infection.
View details for DOI 10.1177/2192568218760540
View details for PubMedID 30443474
View details for PubMedCentralID PMC6232712
Impact of Age on 30-day Complications After Adult Deformity Surgery.
2018; 43 (2): 120-126
A retrospective analysis.The aim of this study was to identify whether age is a risk factor for postoperative complications after adult deformity surgery (ADS).Spinal deformity is a prevalent cause of morbidity in the elderly population, occurring in as many as 68% of patients older than 60 years. Given the increasing prevalence of adult spinal deformities and an aging population, understanding the safety of ADS in elderly patients is becoming increasingly important.A retrospective cohort analysis was performed on the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database from 2010 to 2014. Patients (≥18 years of age) from the NSQIP database undergoing ADS were separated into age-based cohorts (≤52, 53-61, 62-69, and ≥70 years of age). Age groups were determined by interquartile analysis. Chi-squared, t tests, and multivariate logistic regression models were used to identify independent risk factors.A total of 5805 patients met the inclusion criteria. Age groups 1, 2, 3, and 4 contained 1518 (26.1%), 1478 (25.4%), 1451 (25.0%), and 1358 (23.4%) patients, respectively. Multivariate logistic regression analysis revealed increasing age (relative to age group 1) to be an independent risk factor for prolonged length of stay [odds ratio (OR) 1.39, confidence interval (CI) 1.12-1.69], all complications (OR 1.64, CI 1.35-2.00), renal complications (OR 3.45, CI 1.43-8.33), urinary tract infection (OR 2.70, CI 1.49-4.76), postoperative transfusion (OR 1.47, CI 1.20-1.82), and unplanned readmission (OR 1.64, CI 1.18-2.23). Gradations in ORs existed between the different cohorts, such that the deleterious effect of age was less pronounced in cohort 3 compared with cohort 4, and even more less so between cohort 2 and cohort 4.Age has been shown to be an independent risk factor for increased length of stay, all complications, renal complications, urinary tract infection, transfusion, and unplanned readmission.3.
View details for DOI 10.1097/BRS.0000000000001832
View details for PubMedID 27488301
Diabetes Mellitus as a Risk Factor for Acute Postoperative Complications Following Elective Adult Spinal Deformity Surgery.
Global spine journal
2018; 8 (6): 615-621
Retrospective cohort study.Diabetes mellitus is a highly prevalent disease in the United States. Adult spinal deformity (ASD) surgery encompasses a wide variety of spinal disorders and is associated with a morbidity rate between 20% and 80%. Considering utilization of spinal surgery will continue to increase, this study investigates the influence of diabetes mellitus on acute postoperative outcomes following elective ASD surgery.The 2010-2014 American College of Surgeon's National Surgical Quality Improvement Program database was queried using Current Procedural Terminology and International Classification of Diseases (9th Revision) diagnosis codes relevant to ASD surgery. Patients were divided into cohorts based on their diabetic status. Bivariate and multivariate logistic regression analyses were employed to identify which 30-day postoperative outcomes patients are at risk for.A total of 5809 patients met the inclusion criteria for the study of which 4553 (84.2%) patients were nondiabetic, 578 (10.7%) patients had non-insulin-dependent diabetes mellitus (NIDDM), and 275 (5.1%) patients had insulin-dependent diabetes mellitus (IDDM). Diabetes status was significantly associated with length of stay ≥5 days (NIDDM: odds ratio [OR] = 1.27, 95% confidence interval [CI] = 1.02-1.58, P = .034; IDDM: OR = 1.55, 95% CI = 1.15-2.09, P = .004), any complication (NIDDM: OR = 1.26, 95% CI = 1.01-1.58, P = .037), urinary tract infection (NIDDM: OR = 1.87, 95% CI = 1.14-3.05, P = .012), and cardiac complications (IDDM: OR = 4.05, 95% CI = 1.72-9.51, P = .001).Given the prevalence of diabetes, surgeons will invariably encounter these patients for ASD surgery. The present study identifies the increased risk NIDDM and IDDM patients experience following ASD surgery. Quantification of this increased risk may improve the selection of appropriate surgical candidates, patient risk stratification, and patient postoperative safety.
View details for DOI 10.1177/2192568218761361
View details for PubMedID 30202716
View details for PubMedCentralID PMC6125929
Predictors for Non-Home Patient Discharge Following Elective Adult Spinal Deformity Surgery.
Global spine journal
2018; 8 (3): 266-272
Retrospective cohort study.Adult spinal deformity (ASD) surgery encompasses a wide variety of spinal disorders and is associated with a morbidity rate between 20% and 80%. The utilization of spinal surgery has increased and this trend is expected to continue. To effectively deal with an increasing patient volume, identifying variables associated with patient discharge destination can expedite placement and reduce length of stay.The 2013-2014 American College of Surgeons National Surgical Quality Improvement Program database was queried using Current Procedural Terminology and International Classification of Diseases, Ninth Revision diagnosis codes relevant to ASD. Patients were divided based on discharge destination. Bivariate and multivariate logistic regression analyses were employed to identify predictors for patient discharge destination and hospital length of stay.A total of 4552 patients met inclusion criteria, of which 1102 (24.2%) had non-home discharge. Multivariate regression revealed total relative value unit (odds ratio [OR] = 1.01, 95% confidence interval [CI] = 1.00-1.01); female sex (OR = 1.54, 95% CI = 1.32-1.81); American Indian, Alaska Native, Asian, Native Hawaiian, or Pacific Islander versus black race (OR = 0.52, 95% CI = 0.35-0.78, P = .002); age ≥65 years (OR = 3.72, 95% CI = 3.19-4.35); obesity (OR = 1.18, 95% CI = 1.01-1.38, P = .034); partially/totally functionally dependent (OR = 2.11, 95% CI = 1.49-2.99); osteotomy (OR = 1.42, 95% CI = 1.12-1.80, P = .004) pelvis fixation (OR = 2.38, 95% CI = 1.82-3.11); operation time ≥4 hours (OR = 1.74, 95% CI = 1.47-2.05); recent weight loss (OR = 7.66, 95% CI = 1.52-38.65; P = .014); and American Society of Anesthesiologists class ≥3 (OR = 1.80, 95% CI = 1.53-2.11) as predictors of non-home discharge. P values were <.001 unless otherwise noted. Additionally, multivariate regression found non-home discharge to be a significant variable in prolonged length of stay.The authors suggest these results can be used to inform patients preoperatively of expected discharge destination, anticipate patient discharge needs postoperatively, and reduce health care costs and morbidity associated with prolonged LOS.
View details for DOI 10.1177/2192568217717971
View details for PubMedID 29796375
View details for PubMedCentralID PMC5958482
Impact of Preoperative Anemia on Outcomes in Adults Undergoing Elective Posterior Cervical Fusion.
Global spine journal
2017; 7 (8): 787-793
Retrospective analysis of prospectively collected data.Few studies have investigated the role of preoperative anemia on postoperative outcomes of posterior cervical fusion. This study looked to investigate the potential relationship between preoperative anemia and postoperative outcomes following posterior cervical spine fusion.Data from patients undergoing elective posterior cervical fusions between 2005 and 2012 was collected from the American College of Surgeons National Surgical Quality Improvement Program database using inclusion/exclusion criteria. Multivariate analyses were used to identify the predictive power of anemia for postoperative outcomes.A total of 473 adult patients undergoing elective posterior cervical fusions were identified with 106 (22.4%) diagnosed with anemia preoperatively. Anemic patients had higher rates of diabetes (P = .0001), American Society of Anesthesiologists scores ≥3 (P < .0001), and higher dependent functional status prior to surgery (P < .0001). Intraoperatively, anemic patients also had higher rates of neuromuscular injuries (P = .0303), stroke (P = .013), bleeding disorders (P = .0056), lower albumin (P < .0001), lower hematocrit (P < .0001), and higher international normalized ratio (P = .002). Postoperatively, anemic patients had higher rates of complications (P < .0001), death (P = .008), blood transfusion (P = .001), reoperation (P = .012), unplanned readmission (P = .022), and extended length of stay (>5 days; P < .0001).Preoperative anemia is linked to a number of postoperative complications, which can increase length of hospital stay and increase the likelihood of reoperation. Identifying preoperative anemia may play a role in optimizing and minimizing the complication rates and severity of comorbidities following posterior cervical fusion.
View details for DOI 10.1177/2192568217705654
View details for PubMedID 29238644
View details for PubMedCentralID PMC5722000
Hospital-Acquired Conditions in Adult Spinal Deformity Surgery: Predictors for Hospital-Acquired Conditions and Other 30-Day Postoperative Outcomes.
2017; 42 (8): 595-602
A retrospective study of prospectively collected data.The aim of this study was to identify risk factors in developing hospital-acquired conditions (HACs) and association of HACs with other 30-day complications in the adult spinal deformity (ASD) population.HACs are subject to a nonpayment policy by the Center for Medicare and Medicaid Services and provide an incentive for medical institutions to improve patient safety. HACs in the ASD population may further exacerbate the already high rates of postoperative morbidity and mortality.The 2010 to 2014 ACS-NSQIP database was queried using Current Procedural Terminology (CPT) codes for adults who had fusion for spinal deformity. Patients were divided into two cohorts on the basis of the development of an HAC, as defined as a case of surgical site infection, urinary tract infection, or venous thromboembolism. Univariate and multivariate logistic regression analyses were employed to determine predictors for HACs and association of HACs with other 30-day postoperative outcomes.Five thousand eight hundred nineteen patients met the inclusion criteria for the study of whom 313 (5.4%) had an HAC. Multivariate logistic regression analysis revealed that age 61 to 70 versus ≤50 years [odds ratio (OR) = 1.58, 1.10-2.27, P = 0.013], 71 to 80 versus ≤50 years (OR = 1.94, 1.31-2.87, P = 0.001), and >80 versus ≤50 years (OR = 2.30, 1.21-4.37, P = 0.011), dependent/partially dependent versus independent functional status (OR = 1.74, 1.13-2.68, P = 0.011), combined versus anterior surgical approach (OR = 2.46, 1.43-4.24, P = 0.001), and posterior versus anterior surgical approach (OR = 1.64, 1.19-2.25, P = 0.002), osteotomies (OR = 1.61, 1.22-2.13, P = 0.001), steroid use (OR = 2.19, 1.39-3.45, P = 0.001), obesity (OR = 1.38, 1.09-1.74, P = 0.007), and operation time ≥4 hours (OR = 2.42, 1.82-3.21, P < 0.001) were predictive factors in developing an HAC.Several modifiable and nonmodifiable factors (age, functional status, surgical approach, utilization of osteotomies, steroid use, obesity, and operation time ≥4 hours) were associated with developing an HAC. HACs were also risk factors for other postoperative complications.3.
View details for DOI 10.1097/BRS.0000000000001840
View details for PubMedID 27496667
Comparing National Inpatient Sample and National Surgical Quality Improvement Program: An Independent Risk Factor Analysis for Risk Stratification in Anterior Cervical Discectomy and Fusion.
2017; 42 (8): 565-572
Retrospective study of prospectively collected data.To explore interdatabase reliability between National Inpatient Sample (NIS) and National Surgical Quality Improvement Program (NSQIP) for anterior cervical discectomy and fusion (ACDF) in data collection and its impact on subsequent statistical analyses.Clinical studies in orthopedics using national databases are ubiquitous, but analytical differences across databases are largely unexplored.A retrospective cohort study of patients undergoing ACDF surgery was performed in NIS and NSQIP. Key demographic variables, comorbidities, intraoperative characteristics, and postoperative complications were analyzed via bivariate and multivariate analyses.A total of 112,162 patients were identified from NIS and 10,617 from NSQIP. Bivariate analysis revealed small, but significant, differences between patient demographics, whereas patient comorbidities and ACDF intraoperative variables were largely much more distinct across the two databases. Multivariate analysis identified independent risk factors between NIS and NSQIP for mortality, cardiac complications, and postoperative sepsis, some of which were identified in both but most of which were unique to one database. Identification of independent risk factors from both databases specifically highlights their greater validity and importance in stratifying patient risks. In addition, NSQIP was found to be a more accurate predictor for complications based on the average areas under the receiver-operating curve (CNSQIP = 0.83 vs. CNIS = 0.81) across the multivariate models. Complication rate analysis between inpatient and outpatient settings in NSQIP showed the importance of at least 30-day patient follow up, which was devoid in NIS data tabulation and further marked its weakness compared with NSQIP.Despite having largely similar patient demographics, this study highlights critical risk factors for ACDF and demonstrates how different patient profiles can be across NIS and NSQIP, the impact of such differences on identification of independent risk factors, and how NSQIP is ultimately better suited for adverse-event studies.3.
View details for DOI 10.1097/BRS.0000000000001850
View details for PubMedID 27513227
High-Risk Subgroup Membership Is a Predictor of 30-Day Morbidity Following Anterior Lumbar Fusion.
Global spine journal
2017; 7 (8): 762-769
Retrospective cohort study.To determine if membership in a high-risk subgroup is predictive of morbidity and mortality following anterior lumbar fusion (ALF).The American College of Surgeons National Surgical Quality Improvement Program database was utilized to identify patients undergoing ALF between 2010 and 2014. Multivariate analysis was utilized to identify high-risk subgroup membership as an independent predictor of postoperative complications.Members of the elderly (≥65 years) (OR = 1.3, P = .02) and non-Caucasian (black, Hispanic, other) (OR = 1.7, P < .0001) subgroups were at greater risk for a LOS ≥5 days. Obese patients (≥30 kg/m2 ) were at greater risk for an operative time ≥4 hours (OR = 1.3, P = .005), and wound complications (OR = 1.8, P = .024) compared with nonobese patients. Emergent procedures had a significantly increased risk for LOS ≥5 days (OR = 4.9, P = .021), sepsis (OR = 14.8, P = .018), and reoperation (OR = 13.4, P < .0001) compared with nonemergent procedures. Disseminated cancer was an independent risk factor for operative time ≥4 hours (OR = 8.4, P < .0001), LOS ≥5 days (OR = 15.2, P < .0001), pulmonary complications (OR = 7.4, P = .019), and postoperative blood transfusion (OR = 3.1, P = .040).High-risk subgroup membership is an independent risk factor for morbidity following ALF. These groups should be targets for aggressive preoperative optimization, and quality improvement initiatives.
View details for DOI 10.1177/2192568217696691
View details for PubMedID 29238640
View details for PubMedCentralID PMC5721989
Anesthesia Duration as an Independent Risk Factor for Early Postoperative Complications in Adults Undergoing Elective ACDF.
Global spine journal
2017; 7 (8): 727-734
Retrospective study.To determine the presence of any potential associations between anesthesia time with postoperative outcome and complications following elective anterior cervical discectomy and fusion (ACDF).Patients who underwent elective ACDF were identified in the American College of Surgeons National Quality Improvement Program database. Patient demographics, medical comorbidities, and perioperative and postoperative complications up to 30 days were analyzed by univariate and multivariate analysis.A total of 3801 patients undergoing elective ACDF were identified. Patients were subdivided into quintiles of anesthesia time: Group 1, 48 to 129 minutes (n = 761, 20%); Group 2, 129 to 156 minutes (n = 760, 20%); Group 3, 156 to 190 minutes (n = 760, 20%); Group 4, 190 to 245 minutes (n = 760, 20%); and Group 5, 245 to 1025 minutes (n = 760, 20%). Univariate analysis showed significantly higher rates of any complication (P < .0001), pulmonary complication (P < .0001), intra-/postoperative blood transfusions (P < .0001), sepsis (P = .017), wound complications (P = .002), total length of stay >5 days (P < .0001), and return to operating room (P = .006) in the highest quintile compared to those of other groups. Multivariate regression analysis revealed that prolonged anesthesia was an independent factor for increased odds of overall complications (odds ratio [OR] = 2.71, P = .012), venous thromboembolism (OR = 2.69, P = .011), and return to the operating room (OR = 2.92, P = .004). The 2 groups with the longest anesthesia durations (quintiles 4 and 5) had increased total length of stay more than 5 days (for quintile 4, OR = 3.10, P = .0004; for quintile 5, OR = 3.61, P < .0001).Prolonged anesthesia duration is associated with increased odds of complication, venous thromboembolism, increased length of stay, and return to the operating room.
View details for DOI 10.1177/2192568217701105
View details for PubMedID 29238635
View details for PubMedCentralID PMC5721997
ASA Classification as a Risk Stratification Tool in Adult Spinal Deformity Surgery: A Study of 5805 Patients.
Global spine journal
2017; 7 (8): 719-726
Retrospective analysis of prospectively collected data.Adult spinal deformity (ASD) surgery is a highly complex procedure that has high complication rates. Risk stratification tools can improve patient management and may lower complication rates and associated costs. The goal of this study was to identify the independent association between American Society of Anesthesiologists (ASA) class and postoperative outcomes following ASD surgery.The 2010-2014 American College of Surgeons National Surgical Quality Improvement Program database was queried using Current Procedural Terminology and International Classification of Diseases, Ninth Revision, codes relevant to ASD surgery. Patients were divided based on their ASA classification. Bivariate and multivariate logistic regression analyses were employed to quantify the increased risk of 30-day postoperative complications for patients with increased ASA scores.A total of 5805 patients met the inclusion criteria, 2718 (46.8%) of which were ASA class I-II and 3087 (53.2%) were ASA class III-IV. Multivariate logistic regression revealed ASA class to be a significant risk factor for mortality (odds ratio [OR] = 21.0), reoperation within 30 days (OR = 1.6), length of stay ≥5 days (OR = 1.7), overall morbidity (OR = 1.4), wound complications (OR = 1.8), pulmonary complications (OR = 2.3), cardiac complications (OR = 3.7), intra-/postoperative red blood cell transfusion (OR = 1.3), postoperative sepsis (OR = 2.7), and urinary tract infection (OR = 1.6).This is the first study evaluating the role of ASA class in ASD surgery with a large patient database. Use of ASA class as a metric for preoperative health was verified and the association of ASA class with postoperative morbidity and mortality in ASD surgery suggests its utility in refining the risk stratification profile and improving preoperative patient counseling for those individuals undergoing ASD surgery.
View details for DOI 10.1177/2192568217700106
View details for PubMedID 29238634
View details for PubMedCentralID PMC5721995
Outcomes of Short Fusion versus Long Fusion for Adult Degenerative Scoliosis: A Systematic Review and Meta-analysis.
2017; 9 (4): 342-349
The objective of this study was to evaluate differences in clinical and radiographic outcomes between short (<3 levels) and long (≥3 levels) fusions in the setting of degenerative lumbar scoliosis. A literature search was performed from six electronic databases. The key terms of "degenerative scoliosis" OR "lumbar scoliosis" AND "fusion" were combined and used as MeSH subheadings. From relevant studies identified, demographic data, complication rates, Oswestry Disability Index (ODI), and radiographic parameters were extracted and the data was pooled and analyzed. Long fusion was associated with comparable overall complication rates to short fusion (17% vs 14%, P = 0.20). There was a significant difference in the incidence of pulmonary complications when comparing short versus long fusion (0.42% vs 2.70%; P = 0.02). No significant difference was found in terms of motor, sensory complications, infections, construct-related or cardiac complications, pseudoarthrosis, dural tears, cerebrospinal fluid (CSF) leak, or urinary retention. A longer fusion was associated with a greater reduction in coronal Cobb angle and increases in lumbar lordosis, but both findings failed to achieve statistical significance. The ODI was comparable across both cohorts. If shorter fusion lengths are clinically indicated, they should be used instead of longer fusion lengths to reduce perioperative time, costs, and some other complications. However, there are no statistically significant differences in terms of radiographically measurable restoration associated with a short or long fusion.
View details for DOI 10.1111/os.12357
View details for PubMedID 29178306
View details for PubMedCentralID PMC6584300
Early Complications and Outcomes in Adult Spinal Deformity Surgery: An NSQIP Study Based on 5803 Patients.
Global spine journal
2017; 7 (5): 432-440
Retrospective analysis.The purpose of this study is to determine the incidence, impact, and risk factors for short-term postoperative complications following elective adult spinal deformity (ASD) surgery.Current Procedural Terminology codes were used to query the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) for adults who underwent spinal deformity surgery from 2010 to 2014. Patients were separated into groups of those with and without complications. Univariate analysis and multivariate logistic regression were used to assess the impact of patient characteristics and operative features on postoperative outcomes.In total, 5803 patients were identified as having undergone ASD surgery in the NSQIP database. The average patient age was 59.5 (±13.5) years, 59.0% were female, and 81.1% were of Caucasian race. The mean body mass index was 29.5(±6.6), with 41.9% of patients having a body mass index of 30 or higher. The most common comorbidities were hypertension requiring medication (54.5%), chronic obstructive pulmonary disease (4.9%), and bleeding disorders (1.2%). Nearly a half of the ASD patients had an operative time >4 hours. The posterior fusion approach was more common (56.9%) than an anterior one (39.6%). The mean total relative value unit was 73.4 (±28.8). Based on multivariate analyses, several patient and operative characteristics were found to be predictive of morbidity.Surgical correction of ASD is associated with substantial risk of intraoperative and postoperative complications. Preoperative and intraoperative variables were associated with increased morbidity and mortality. This data may assist in developing future quality improvement activities and saving costs through measurable improvement in patient safety.
View details for DOI 10.1177/2192568217699384
View details for PubMedID 28811987
View details for PubMedCentralID PMC5544158
Elderly Age as a Risk Factor for 30-Day Postoperative Outcomes Following Elective Anterior Cervical Discectomy and Fusion.
Global spine journal
2017; 7 (5): 425-431
Retrospective analysis of prospectively collected data.Anterior cervical discectomy and fusion (ACDF) is one of the most commonly performed spinal procedures. Considering the high success and low complications rate of ACDF and high prevalence of age-related degeneration of the cervical spine, the rates of ACDF are expected to continually rise. The objective is to identify the association between patient age and 30-day postoperative outcomes following elective ACDF.The 2010-2014 ACS-NSQIP database was queried using Current Procedural Terminology (CPT) codes 22551 or 22554. Patients were divided into age quartiles (18-45, 46-52, 53-60, and ≥61 years). Bivariate and multivariate logistic regression analyses were employed to quantify the increased risk of 30-day postoperative complications in the elderly patient population.A total of 20 563 patients met the inclusion criteria for the study. The analyses found quartile 4 had an increased odds of length of stay (LOS) ≥5 days (odds ratio [OR] = 2.05, confidence interval [CI ] = 1.62-2.60), pulmonary complications (OR = 3.25, CI = 1.81-5.84), urinary tract infections (UTI) (OR = 2.25, 1.04-4.87, P = .038), cardiac complication (OR = 6.01, CI = 1.36-26.62, P = .018), and sepsis (OR = 4.38, CI = 1.30-14.70, P = .017). Quartiles 2 and 4 had an increased odds of venous thromboembolism (OR = 3.13, CI = 1.14-8.56, P = .026; OR = 3.83, CI = 1.44-10.20, P = .007). Quartiles 3 and 4 experienced an increased odds of unplanned readmission (OR = 1.44, CI = 1.01-2.05, P = .045; OR = 1.88, CI = 1.33-2.66). All P values are <.001 unless otherwise noted.Elderly patients experienced an increased odds of LOS ≥5 days, pulmonary complications, cardiac compilations, venous thromboembolism, UTI, sepsis, and unplanned readmission. Identification of these factors can improve the selection of appropriate surgical candidates and postoperative safety.
View details for DOI 10.1177/2192568217699383
View details for PubMedID 28811986
View details for PubMedCentralID PMC5544157
Intramedullary Nail Fixation of Atypical Femur Fractures With Bone Marrow Aspirate Concentrate Leads to Faster Union: A Case-Control Study.
Journal of orthopaedic trauma
2017; 31 (7): 358-362
To evaluate bone marrow aspirate concentrate (BMAC) use in the treatment of AFF.Retrospective case control.Level 1 trauma center.Complete AFF, defined according to American Society of Bone and Mineral Research (ASBMR) criteria, from September 2009 to April 2015 with minimum 1-year follow-up.Operative treatment with antegrade intramedullary nails. Beginning June 2014, BMAC from the ipsilateral iliac crest was added to all AFFs.Time to union as determined by a blinded panel of 3 attending orthopaedic surgeons, union rates, complications.Thirty-five patients with 36 AFFs were reviewed, of which 33 AFFs were included and 11 received BMAC. Alendronate was the most commonly prescribed bisphosphonate, with a similar mean duration of use in controls and BMAC cases (5.6 versus 6 years, P = 0.79). BMAC use significantly decreased time to union (3.5 versus 6.8 months, P = 0.004). Varus malreduction was associated with a significant delay in union (9.7 versus 4.7 months, P = 0.04). Overall, 1 year union rate was 86.2% and nonsignificantly higher in BMAC compared with controls (100.0% versus 77.3%, P = 0.11). Multivariate analysis revealed BMAC and varus malreduction as independent predictors of time to union. There were no complications related to BMAC use.Our findings support intramedullary nailing of AFFs as an effective treatment option with a low surgical complication rate and highlight the importance of avoiding varus malreduction. BMAC use significantly reduced time to fracture union without an increase in surgical complication rates.Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.
View details for DOI 10.1097/BOT.0000000000000851
View details for PubMedID 28632656
Bone morphogenetic protein use in spine surgery in the United States: how have we responded to the warnings?
The spine journal : official journal of the North American Spine Society
2017; 17 (9): 1247-1254
Recombinant human bone morphogenetic protein-2 (rhBMP-2) has been widely adopted as a fusion adjunct in spine surgery since its approval in 2002. A number of concerns regarding adverse effects and potentially devastating complications of rhBMP-2 use led to a Food and Drug Administration (FDA) advisory issued in 2008 cautioning its use, and a separate warning about its potential complications was published by The Spine Journal in 2011.To compare trends of rhBMP-2 use in spine surgery after the FDA advisory in 2008 and The Spine Journal warning in 2011.Retrospective cross-sectional study using a national database.All patients from 2002 to 2013 who underwent spinal fusion surgery at an institution participating in the Nationwide Inpatient Sample (NIS).Proportion of spinal fusion surgeries using rhBMP-2.We queried the NIS from 2002 to 2013 and used International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure codes to identify spinal fusion procedures and those that used rhBMP-2. Procedures were subdivided into primary and revision fusions, and by region of the spine. Cervical and lumbosacral fusions were further stratified into anterior and posterior approaches. The percentage of cases using BMP was plotted across time. A linear regression was fit to the data from quarter 3 of 2008 (FDA advisory) through quarter 1 of 2011, and a separate regression was fit to the data from quarter 2 of 2011 (The Spine Journal warning) onward. The slopes of these regression lines were statistically compared to determine differences in trends. No funding was received to conduct this study, and no authors had any relevant conflicts of interest.A total of 4,167,079 patients in the NIS underwent spinal fusion between 2002 and 2013. We found a greater decrease in rhBMP-2 use after The Spine Journal warning compared with the FDA advisory for all fusion procedures (p=.006), primary fusions (p=.006), and revision fusions (p=.004). Lumbosacral procedures also experienced a larger decline in rhBMP-2 use after The Spine Journal article as compared with the FDA warning (p=.0008). This pattern was observed for both anterior and posterior lumbosacral fusions (p≤.0001 for both). Anterior cervical fusion was the only procedure that demonstrated a decline in rhBMP-2 use after the FDA advisory that was statistically greater than after The Spine Journal article (p=.02).Warnings sanctioned through the spine literature may have a greater influence on practice of the spine surgery community as compared with advisories issued by the FDA.Comprehensive guidelines regarding safe and effective use of rhBMP-2 must be established.
View details for DOI 10.1016/j.spinee.2017.04.030
View details for PubMedID 28456674
Thirty-Day Morbidity Associated with Pelvic Fixation in Adult Patients Undergoing Fusion for Spinal Deformity: A Propensity-Matched Analysis.
Global spine journal
2017; 7 (1): 39-46
Retrospective study of prospectively collected data.To determine if patients undergoing spinal deformity surgery with pelvic fixation are at an increased risk of morbidity.The American College of Surgeons National Surgical Quality Improvement Program is a large multicenter clinical registry that prospectively collects preoperative risk factors, intraoperative variables, and 30-day postoperative morbidity and mortality outcomes from ~400 hospitals nationwide. Current Procedural Terminology codes were used to query the database between 2010 and 2014 for adults who underwent fusion for spinal deformity. Patients were separated into groups of those with and without pelvic fixation. Univariate analysis and multivariate logistic regression were used to analyze the effect of pelvic fixation on the incidence of postoperative morbidity and other surgical outcomes.Multivariate analysis showed that pelvic fixation was a significant predictor of overall morbidity (odds ratio [OR] = 2.3, 95% confidence interval [CI]: 1.7 to 3.1, p = 0.0002), intra- or postoperative blood transfusion (OR = 2.3, 95% CI: 1.7 to 3.1 p < 0.0001), extended operative time (OR = 4.7, 95% CI: 3.1 to 7.0 p < 0.0001), and length of stay > 5 days (OR = 2.1, 95% CI 1.5 to 2.8, p < 0.0001) in patients undergoing fusion for spinal deformity. However, fusion to the pelvis did not lead to additional risk for other complications, including wound complications (p = 0.3191).Adult patients undergoing spinal deformity surgery with pelvic fixation were not susceptible to increased morbidity beyond increased blood loss, greater operative time, and extended length of stay.
View details for DOI 10.1055/s-0036-1583946
View details for PubMedID 28451508
View details for PubMedCentralID PMC5400170
Predictors for Patient Discharge Destination After Elective Anterior Cervical Discectomy and Fusion.
2017; 42 (20): 1538-1544
Retrospective study of prospectively collected data.To identify risk factors for nonhome patient discharge after elective anterior cervical discectomy and fusion (ACDF).ACDF is one of the most performed spinal procedures and this is expected to increase in the coming years. To effectively deal with an increasing patient volume, identifying variables associated with patient discharge destination can expedite placement applications and subsequently reduce hospital length of stay.The 2011 to 2014 ACS-NSQIP database was queried using Current Procedural Terminology (CPT) codes 22551 or 22554. Patients were divided into two cohorts based on discharge destination. Bivariate and multivariate logistic regression analyses were employed to identify predictors for patient discharge destination and extended hospital length of stay.A total of 14,602 patients met the inclusion criteria for the study of which 498 (3.4%) had nonhome discharge. Multivariate logistic regression found that Hispanic versus Black race/ethnicity (odds ratio, OR =0.21, 0.05-0.91, P =0.037), American Indian or Alaska Native, Asian, Native Hawaiian or Pacific Islander versus Black race/ethnicity (OR = 0.52, 0.34-0.80, p-value = 0.003), White versus Black race/ethnicity (OR = 0.55, 0.42-0.71), elderly age ≥65 years (OR = 3.32, 2.72-4.06), obesity (OR = 0.77, 0.63-0.93, P = 0.008), diabetes (OR = 1.32, 1.06-1.65, P = 0.013), independent versus partially/totally dependent functional status (OR = 0.11, 0.08-0.15), operation time ≥4 hours (OR = 2.46, 1.87-3.25), cardiac comorbidity (OR = 1.38, 1.10-1.72, P = 0.005), and ASA Class ≥3 (OR = 2.57, 2.05-3.20) were predictive factors in patient discharge to a facility other than home. In addition, multivariate logistic regression analysis also found nonhome discharge to be the most predictive variable in prolonged hospital length of stay.Several predictive factors were identified in patient discharge to a facility other than home, many being preoperative variables. Identification of these factors can expedite patient discharge applications and potentially can reduce hospital stay, thereby reducing the risk of hospital acquired conditions and minimizing health care costs.3.
View details for DOI 10.1097/BRS.0000000000002140
View details for PubMedID 28252556
Analysis of Risk Factors for Major Complications Following Elective Posterior Lumbar Fusion.
2017; 42 (17): 1347-1354
Retrospective study of prospectively collected data.To identify risk factors for the development of any major complication after elective posterior lumbar fusion (PLF).PLF is one of the most performed fusion techniques with utilization rates increasing by 356% between 1993 and 2001. Surgical and anesthetic advances have made the option of surgery more accessible for elderly patients with a larger comorbidity burden. Identifying risk factors for the development of major complications after elective PLF is important for patient risk stratification and patient safety efforts.The 2011 to 2014 American College of Surgeon's National Surgical Quality Improvement Program database was queried using Current Procedural Terminology codes 22612, 22630, and 22633. Patients were divided into two cohorts based on the development of any major complication. Bivariate and multivariate logistic regression analyses were employed to identify predictors for the development of ≥ 1, ≥ 2, and ≥ 3 major complications.A total of 7761 patients met the inclusion criteria for the study of which, 2055 (26.5%) patients developed one major complication, 249 (3.2%) patients developed two major complications, and 151 (1.9%) patients developed three major complications. The most common complication was intra/postoperative red blood cell transfusion (23.2%). Three multivariate logistic regression models were employed to identify factors associated with ≥ 1, ≥ 2, and ≥ 3 major complications. Patient variables present across all three models were osteotomy, pelvic fixation, operation time ≥4 hours, bleeding disorder, and American Society of Anesthesiology Class ≥ 3.Several risk factors were identified for the development of major complications after elective PLF. Identification of these factors can improve the selection of appropriate surgical candidates, patient risk stratification, and patient postoperative safety.3.
View details for DOI 10.1097/BRS.0000000000002090
View details for PubMedID 28146019
Frailty is associated with morbidity in adults undergoing elective anterior lumbar interbody fusion (ALIF) surgery.
The spine journal : official journal of the North American Spine Society
2017; 17 (4): 538-544
Prior studies have suggested no significant differences in functional status and postoperative complications of elderly versus nonelderly patients undergoing posterior lumbar interbody fusion; however, similar studies have not been comprehensively investigated in the setting of anterior lumbar interbody fusion (ALIF).The objective was to quantify the ability of the modified Frailty Index (mFI) to predict postoperative events in patients undergoing ALIF.Secondary analysis of prospectively collected data.Patients undergoing ALIF in the National Surgical Quality Improvement Program (NSQIP) participant files for the period 2010 through 2014.Outcome measures included any postoperative complication, return to operating room (OR), and length of stay >5 days.NSQIP participant files from 2010 to 2014 were used to identify patients undergoing ALIF. The mFI used in the present study is an 11-variable assessment that maps 16 NSQIP variables to 11 variables in the Canadian Study of Health and Ageing Frailty Index. Univariate analysis and multivariable logistic regression models were used to compare the relative strength of association between mFI with outcome variables of interest.In total, 3,920 ALIF cases were identified and grouped according to their mFI score: 0 (n=2,025), 0.09 (n=1,382), 0.18 (n=464), or ≥0.27 (n=49). As the mFI increased from 0 (no frailty-associated variables) to 0.27 (4 of 11) or higher, there was a significant stepwise increase in any complication from 10.8% to 32.7%. After multivariable regression analysis, no significant association was found between higher mFI scores with urinary tract infections and venous thromboembolism. High frailty scores were significant predictors of any complication (mFI of ≥0.27 [reference: 0]; OR 2.4; p=.040) and pulmonary complications (mFI score ≥0.27; OR 7.5; p=.001).In summary, high mFI scores were found to be independently associated with any complication and pulmonary complications in patients who underwent ALIF. The use of mFI together with traditional risk factors may help better identify high-surgical risk patients, which may be useful for preoperative and postoperative care optimization.
View details for DOI 10.1016/j.spinee.2016.10.023
View details for PubMedID 27989724
Drug response consistency in CCLE and CGP.
2016; 540 (7631): E9-E10
View details for DOI 10.1038/nature20580
View details for PubMedID 27905419
View details for PubMedCentralID PMC5554885