Using Predictive Modeling and Machine Learning to Identify Patients Appropriate for Outpatient Anterior Cervical Fusion and Discectomy.
Adult
Ambulatory Surgical Procedures
/ methods
Case-Control Studies
Cervical Vertebrae
/ surgery
Cohort Studies
Databases, Factual
Diskectomy
/ methods
Female
Humans
Machine Learning
Male
Middle Aged
Predictive Value of Tests
Quality Improvement
Retrospective Studies
Risk Factors
Spinal Diseases
/ diagnosis
Spinal Fusion
/ methods
Journal
Spine
ISSN: 1528-1159
Titre abrégé: Spine (Phila Pa 1976)
Pays: United States
ID NLM: 7610646
Informations de publication
Date de publication:
15 May 2021
15 May 2021
Historique:
pubmed:
12
12
2020
medline:
22
6
2021
entrez:
11
12
2020
Statut:
ppublish
Résumé
Retrospective, case-control. The aim of this study was to use predictive modeling and machine learning to develop novel tools for identifying patients who may be appropriate for single-level outpatient anterior cervical fusion and discectomy (ACDF), and to compare these to legacy metrics. ACDF performed in an ambulatory surgical setting has started to gain popularity in recent years. Currently there are no standardized risk-stratification tools for determining which patients may be safe candidates for outpatient ACDF. Adult patients with American Society of Anesthesiologists (ASA) Class 1, 2, or 3 undergoing one-level ACDF in inpatient or outpatient settings were identified in the National Surgical Quality Improvement Program database. Patients were deemed as "unsafe" for outpatient surgery if they suffered any complication within a week of the index operation. Two different methodologies were used to identify unsafe candidates: a novel predictive model derived from multivariable logistic regression of significant risk factors, and an artificial neural network (ANN) using preoperative variables. Both methods were trained using randomly split 70% of the dataset and validated on the remaining 30%. The methods were compared against legacy risk-stratification measures: ASA and Charlson Comorbidity Index (CCI) using area under the curve (AUC) statistic. A total of 12,492 patients who underwent single-level ACDF met the study criteria. Of these, 9.79% (1223) were deemed unsafe for outpatient ACDF given development of a complication within 1 week of the index operation. The five clinical variables that were found to be significant in the multivariable predictive model were: advanced age, low hemoglobin, high international normalized ratio, low albumin, and poor functional status. The predictive model had an AUC of 0.757, which was significantly higher than the AUC of both ASA (0.66; P < 0.001) and CCI (0.60; P < 0.001). The ANN exhibited an AUC of 0.740, which was significantly higher than the AUCs of ASA and CCI (all, P < 0.05), and comparable to that of the predictive model (P > 0.05). Predictive analytics and machine learning can be leveraged to aid in identification of patients who may be safe candidates for single-level outpatient ACDF. Surgeons and perioperative teams may find these tools useful to augment clinical decision-making.Level of Evidence: 3.
Sections du résumé
STUDY DESIGN
METHODS
Retrospective, case-control.
OBJECTIVE
OBJECTIVE
The aim of this study was to use predictive modeling and machine learning to develop novel tools for identifying patients who may be appropriate for single-level outpatient anterior cervical fusion and discectomy (ACDF), and to compare these to legacy metrics.
SUMMARY OF BACKGROUND DATA
BACKGROUND
ACDF performed in an ambulatory surgical setting has started to gain popularity in recent years. Currently there are no standardized risk-stratification tools for determining which patients may be safe candidates for outpatient ACDF.
METHODS
METHODS
Adult patients with American Society of Anesthesiologists (ASA) Class 1, 2, or 3 undergoing one-level ACDF in inpatient or outpatient settings were identified in the National Surgical Quality Improvement Program database. Patients were deemed as "unsafe" for outpatient surgery if they suffered any complication within a week of the index operation. Two different methodologies were used to identify unsafe candidates: a novel predictive model derived from multivariable logistic regression of significant risk factors, and an artificial neural network (ANN) using preoperative variables. Both methods were trained using randomly split 70% of the dataset and validated on the remaining 30%. The methods were compared against legacy risk-stratification measures: ASA and Charlson Comorbidity Index (CCI) using area under the curve (AUC) statistic.
RESULTS
RESULTS
A total of 12,492 patients who underwent single-level ACDF met the study criteria. Of these, 9.79% (1223) were deemed unsafe for outpatient ACDF given development of a complication within 1 week of the index operation. The five clinical variables that were found to be significant in the multivariable predictive model were: advanced age, low hemoglobin, high international normalized ratio, low albumin, and poor functional status. The predictive model had an AUC of 0.757, which was significantly higher than the AUC of both ASA (0.66; P < 0.001) and CCI (0.60; P < 0.001). The ANN exhibited an AUC of 0.740, which was significantly higher than the AUCs of ASA and CCI (all, P < 0.05), and comparable to that of the predictive model (P > 0.05).
CONCLUSION
CONCLUSIONS
Predictive analytics and machine learning can be leveraged to aid in identification of patients who may be safe candidates for single-level outpatient ACDF. Surgeons and perioperative teams may find these tools useful to augment clinical decision-making.Level of Evidence: 3.
Identifiants
pubmed: 33306613
pii: 00007632-202105150-00006
doi: 10.1097/BRS.0000000000003865
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
665-670Informations de copyright
Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.
Références
Kani KK, Chew FS. Anterior cervical discectomy and fusion: review and update for radiologists. Skeletal Radiol 2018; 47:7–17.
Denaro V, Di Martino A. Cervical spine surgery: an historical perspective. Clin Orthop Relat Res 2011; 469:639–648.
Saifi C, Fein AW, Cazzulino A, et al. Trends in resource utilization and rate of cervical disc arthroplasty and anterior cervical discectomy and fusion throughout the United States from 2006 to 2013. Spine J 2018; 18:1022–1029.
Tetreault L, Goldstein CL, Arnold P, et al. Degenerative cervical myelopathy: a spectrum of related disorders affecting the aging spine. Neurosurgery 2015; 77: (suppl 4): S51–S67.
McClelland S, Oren JH, Protopsaltis TS, et al. Outpatient anterior cervical discectomy and fusion: a meta-analysis. J Clin Neurosci 2016; 34:166–168.
Fountas KN, Kapsalaki EZ, Nikolakakos LG, et al. Anterior cervical discectomy and fusion associated complications. Spine (Phila Pa 1976) 2007; 32:2310–2317.
McGirt MJ, Godil SS, Asher AL, et al. Quality analysis of anterior cervical discectomy and fusion in the outpatient versus inpatient setting: analysis of 7288 patients from the NSQIP database. Neurosurg Focus 2015; 39:E9.
Shiloach M, Frencher SK Jr, Steeger JE, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010; 210:6–16.
Hustedt JW, Chung A, Bohl DD. Development of a risk stratification scoring system to predict general surgical complications in hand surgery patients. J Hand Surg Am 2018; 43:641–648.
Bohl DD, Idarraga AJ, Holmes GBJ, et al. Validated risk-stratification system for prediction of early adverse events following open reduction and internal fixation of closed ankle fractures. J Bone Joint Surg Am 2019; 101:1768–1774.
Bohl DD, Maltenfort MG, Huang R, et al. Development and validation of a risk stratification system for pulmonary embolism after elective primary total joint arthroplasty. J Arthroplasty 2016; 31:187–191.
Saklad M. Grading of patients for surgical procedures. Anesthesiology 1941; 2:281–284.
Stavem K, Hoel H, Skjaker SA, et al. Charlson comorbidity index derived from chart review or administrative data: agreement and prediction of mortality in intensive care patients. Clin Epidemiol 2017; 9:311–320.
De Bruijn B. Revisiting the area under the ROC. Stud Health Technol Inform 2011; 169:532–536.
Purger DA, Pendharkar AV, Ho AL, et al. Outpatient vs inpatient anterior cervical discectomy and fusion: a population-level analysis of outcomes and cost. Neurosurgery 2017; 82:454–464.
Di Capua J, Somani S, Kim JS, et al. Elderly age as a risk factor for 30-day postoperative outcomes following elective anterior cervical discectomy and fusion. Glob spine J 2017; 7:425–431.
Phan K, Wang N, Kim JS, et al. Effect of preoperative anemia on the outcomes of anterior cervical discectomy and fusion. Glob Spine J 2017; 7:441–447.
Bovonratwet P, Fu MC, Tyagi V, et al. Incidence, risk factors, and clinical implications of postoperative hematoma requiring reoperation following anterior cervical discectomy and fusion. Spine (Phila Pa 1976) 2019; 44:543–549.
Zuckerman SL, Mummareddy N, Lakomkin N, et al. Identifying the most appropriate ACDF patients for an ambulatory surgery center: a pilot study using inpatient and outpatient hospital data. Clin Spine Surg 2020; doi:10.1097/BSD.0000000000000967.
doi: 10.1097/BSD.0000000000000967
Cuocolo R, Cipullo MB, Stanzione A, et al. Machine learning applications in prostate cancer magnetic resonance imaging. Eur Radiol Exp 2019; 3:35.
Huang Q, Zhang F, Li X. Machine learning in ultrasound computer-aided diagnostic systems: a survey. BioMed Res Int 2018; 2018:5137904.
Obermeyer Z, Emanuel EJ. Predicting the future-big data, machine learning, and clinical medicine. N Engl J Med 2016; 375:1216–1219.
Zhang Z. A gentle introduction to artificial neural networks. Ann Transl Med 2016; 4:370.
Bertolaccini L, Solli P, Pardolesi A, et al. An overview of the use of artificial neural networks in lung cancer research. J Thorac Dis 2017; 9:924–931.
Arvind V, Kim JS, Oermann EK, et al. Predicting surgical complications in adult patients undergoing anterior cervical discectomy and fusion using machine learning. Neurospine 2018; 15:329–337.