1
|
Zang H, Hu A, Xu X, Ren H, Xu L. Development of machine learning models to predict perioperative blood transfusion in hip surgery. BMC Med Inform Decis Mak 2024; 24:158. [PMID: 38840126 PMCID: PMC11155147 DOI: 10.1186/s12911-024-02555-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/21/2023] [Accepted: 05/28/2024] [Indexed: 06/07/2024] Open
Abstract
BACKGROUND Allogeneic Blood transfusion is common in hip surgery but is associated with increased morbidity. Accurate prediction of transfusion risk is necessary for minimizing blood product waste and preoperative decision-making. The study aimed to develop machine learning models for predicting perioperative blood transfusion in hip surgery and identify significant risk factors. METHODS Data of patients undergoing hip surgery between January 2013 and October 2021 in the Peking Union Medical College Hospital were collected to train and test predictive models. The primary outcome was perioperative red blood cell (RBC) transfusion within 72 h of surgery. Fourteen machine learning algorithms were established to predict blood transfusion risk incorporating patient demographic characteristics, preoperative laboratory tests, and surgical information. Discrimination, calibration, and decision curve analysis were used to evaluate machine learning models. SHapley Additive exPlanations (SHAP) was performed to interpret models. RESULTS In this study, 2431 hip surgeries were included. The Ridge Classifier performed the best with an AUC = 0.85 (95% CI, 0.81 to 0.88) and a Brier score = 0.21. Patient-related risk factors included lower preoperative hemoglobin, American Society of Anesthesiologists (ASA) Physical Status > 2, anemia, lower preoperative fibrinogen, and lower preoperative albumin. Surgery-related risk factors included longer operation time, total hip arthroplasty, and autotransfusion. CONCLUSIONS The machine learning model developed in this study achieved high predictive performance using available variables for perioperative blood transfusion in hip surgery. The predictors identified could be helpful for risk stratification, preoperative optimization, and outcomes improvement.
Collapse
Affiliation(s)
- Han Zang
- Department of Anesthesiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, 100730, China
| | - Ai Hu
- Department of Anesthesiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, 100730, China
| | - Xuanqi Xu
- Key Laboratory of High Confidence Software Technologies (Peking University), Ministry of Education, Beijing, 100084, China
- School of Computer Science, Peking University, Beijing, 100084, China
| | - He Ren
- Beijing HealSci Technology Co., Ltd., Beijing, 100176, China
| | - Li Xu
- Department of Anesthesiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, 100730, China.
| |
Collapse
|
2
|
Vrbica K, Hudec J, Hrdy O, Galko M, Horalkova H, Demlova R, Kubelova M, Repko M, Gal R. Effect of Prophylactic Fibrinogen Concentrate In Scoliosis Surgery (EFISS): a study protocol of two-arm, randomised trial. BMJ Open 2023; 13:e071547. [PMID: 37236666 DOI: 10.1136/bmjopen-2022-071547] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 05/28/2023] Open
Abstract
INTRODUCTION Fibrinogen is one of the essential coagulation factors. Preoperative lower plasma fibrinogen level has been associated with higher blood loss. Scoliosis surgery presents a challenge for the anaesthetic team, one of the reasons being blood loss and transfusion management. Recently, the prophylactic fibrinogen administration has been a debated topic in various indications. It has been described for example, in urological or cardiovascular surgery, as well as in paediatrics. This pilot study is focused on verifying the feasibility of potential large randomised trial and verifying the safety of prophylactic fibrinogen administration in paediatric scoliosis surgery. METHODS AND ANALYSIS A total of 32 paediatric patients indicated for scoliosis surgery will be recruited. Participants will be randomised into study groups in a 1:1 allocation ratio. Patients in the intervention group will receive prophylactic single dose of fibrinogen, in addition to standard of care. Patients in the control group will receive standard of care without study medication prior to skin incision. The primary aim is to assess the safety of prophylactic fibrinogen administration during scoliosis surgery in children, the incidence of any adverse events (AEs) and reactions will be monitored during participation in the study. The secondary objective is to investigate the additional safety information, feasibility and efficacy of a prophylactic fibrinogen administration. The incidence of AEs and reactions according to selected adverse events of special interest will be monitored. All collected data will be subjected to statistical analysis according to a separate statistical analysis plan. ETHICS AND DISSEMINATION This trial follows the applicable legislation and requirements for good clinical practice according to the International Conference on Harmonisation E6(R2). All essential trial documents were approved by the relevant ethics committee and national regulatory authority (State Institute for Drug Control) and their potential amendments will be submitted for approval. TRIAL REGISTRATION NUMBER NCT05391412.
Collapse
Affiliation(s)
- Kamil Vrbica
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Brno, Brno, Czech Republic
- Department of Anaesthesiology and Intensive Care Medicine, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Jan Hudec
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Brno, Brno, Czech Republic
- Department of Anaesthesiology and Intensive Care Medicine, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Ondrej Hrdy
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Brno, Brno, Czech Republic
- Department of Anaesthesiology and Intensive Care Medicine, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Michal Galko
- Department of Orthopaedic Surgery, University Hospital Brno, Brno, Czech Republic
- Department of Orthopaedic Surgery, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Hana Horalkova
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Brno, Brno, Czech Republic
- Department of Anaesthesiology and Intensive Care Medicine, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Regina Demlova
- Department of Pharmacology/CZECRIN, Masaryk University Faculty of Medicine, Brno, Czech Republic
| | - Michaela Kubelova
- Department of Pharmacology/CZECRIN, Masaryk University Faculty of Medicine, Brno, Czech Republic
| | - Martin Repko
- Department of Orthopaedic Surgery, University Hospital Brno, Brno, Czech Republic
- Department of Orthopaedic Surgery, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| | - Roman Gal
- Department of Anaesthesiology and Intensive Care Medicine, University Hospital Brno, Brno, Czech Republic
- Department of Anaesthesiology and Intensive Care Medicine, Masaryk University, Faculty of Medicine, Brno, Czech Republic
| |
Collapse
|
3
|
Terada R, Stewart K, Hansen S, Shettar SS, Butt A, Vallurupalli S, Martin M, Tanaka K. Temporal changes and predictors of perioperative blood transfusion in elective spine surgery: a retrospective analysis of a large surgical database. Br J Anaesth 2023; 130:421-429. [PMID: 36586804 DOI: 10.1016/j.bja.2022.11.020] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/12/2022] [Revised: 11/09/2022] [Accepted: 11/13/2022] [Indexed: 12/31/2022] Open
Abstract
BACKGROUND Allogeneic blood transfusion used to be common in spine surgery. Patient blood management has been widely adopted, and it is important to reassess transfusion predictors in contemporary practice. METHODS A retrospective study of inpatient spine surgery was performed using National Surgical Quality Improvement Program (NSQIP) data from 2011 to 2019. The primary outcome was perioperative transfusion within 72 h of surgery. Multivariable logistic regression and recursive partitioning were used to assess up to 15 variables including patient and surgical data, surgical specialty (orthopaedic surgery vs neurosurgery), and year of surgery. RESULTS The study population included 251 971 US surgical patients; 6.9% of these patients received perioperative blood transfusion. Perioperative transfusions declined over time with the steepest decline from 2011 to 2015. The greatest reduction was seen among orthopaedic cases where the transfusion rate declined from 16.0% to 8.7% between 2011 and 2015. Eight variables were predictive factors in a reduced model: operative time, preoperative haemoglobin, vertebral level, number of vertebral levels, older age, surgeon specialty, arthrodesis, and year of surgery (area under the curve [AUC]=0.880; 95% confidence interval [CI], 0.878-0.883). Overall, longer operative time (>144 min) and greater numbers of vertebral levels had greater associations with transfusion than surgical specialty after adjustments. Prevalence of anaemia (15%) has not substantially declined. CONCLUSIONS Perioperative blood transfusion rate in spine surgery has declined over the past decade. The extent and duration of surgery and preoperative haemoglobin level remain important factors associated with increased odds for perioperative blood transfusion.
Collapse
Affiliation(s)
- Rui Terada
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Kenneth Stewart
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA; Department of Surgery, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Sandra Hansen
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Shashank S Shettar
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Amir Butt
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Santaram Vallurupalli
- Department of Orthopedic Surgery, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Michael Martin
- Department of Neurosurgery, University of Oklahoma Health Science Center, Oklahoma City, OK, USA
| | - Kenichi Tanaka
- Department of Anaesthesiology, University of Oklahoma Health Science Center, Oklahoma City, OK, USA.
| |
Collapse
|
4
|
Pagnussatt Neto E, Lopes da Costa PD, Gurgel SJT, Schmidt Azevedo P, Modolo NSP, do Nascimento Junior P. Plasma Fibrinogen as a Predictor of Perioperative-Blood-Component Transfusion in Major-Nontraumatic-Orthopedic-Surgery Patients: A Cohort Study. Diagnostics (Basel) 2023; 13:diagnostics13050976. [PMID: 36900120 PMCID: PMC10001368 DOI: 10.3390/diagnostics13050976] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/18/2023] [Revised: 02/08/2023] [Accepted: 02/27/2023] [Indexed: 03/08/2023] Open
Abstract
There is a trend towards increased perioperative bleeding in patients with plasma fibrinogen levels < 200 mg/dL-1. This study aimed to assess whether there is an association between preoperative fibrinogen levels and perioperative blood-product transfusion up to 48 h after major orthopedic surgery. This cohort study included 195 patients who underwent primary or revision hip arthroplasty for nontraumatic etiologies. Plasma fibrinogen, blood count, coagulation tests, and platelet count were measured preoperatively. A plasma fibrinogen level of 200 mg/dL-1 was the cutoff value used to predict blood transfusion. The mean (SD) plasma fibrinogen level was 325 (83) mg/dL-1. Only thirteen patients had levels < 200 mg/dL-1, and only one of them received a blood transfusion, with an absolute risk of 7.69% (1/13; 95%CI: 1.37-33.31%). Preoperative plasma fibrinogen levels were not associated with the need for blood transfusion (p = 0.745). The sensitivity and the positive predictive value of plasma fibrinogen < 200 mg/dL-1 as a predictor of blood transfusion were 4.17% (95%CI: 0.11-21.12%) and 7.69% (95%CI: 1.12-37.99%), respectively. Test accuracy was 82.05% (95%CI: 75.93-87.17%), but positive and negative likelihood ratios were poor. Therefore, preoperative plasma fibrinogen level in hip-arthroplasty patients was not associated with the need for blood-product transfusion.
Collapse
Affiliation(s)
- Eugenio Pagnussatt Neto
- Department of Surgical Specialties and Anesthesiology, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
- São Vicente de Paulo Hospital, Passo Fundo 99010-112, RS, Brazil
- Correspondence: ; Tel.: +55-(54)-99975-7010
| | - Paula Daniele Lopes da Costa
- Department of Surgical Specialties and Anesthesiology, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
| | - Sanderland J. Tavares Gurgel
- Department of Surgical Specialties and Anesthesiology, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
| | - Paula Schmidt Azevedo
- Department of Internal Medicine, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
| | - Norma S. Pinheiro Modolo
- Department of Surgical Specialties and Anesthesiology, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
| | - Paulo do Nascimento Junior
- Department of Surgical Specialties and Anesthesiology, School of Medicine, Universidade Estadual Paulista (UNESP), Botucatu 18618-970, SP, Brazil
| |
Collapse
|
5
|
Fibrinogen prophylaxis for reducing perioperative bleeding in patients undergoing radical cystectomy: A double-blind placebo-controlled randomized trial. J Clin Anesth 2021; 73:110373. [PMID: 34098395 DOI: 10.1016/j.jclinane.2021.110373] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2021] [Revised: 04/28/2021] [Accepted: 04/30/2021] [Indexed: 11/22/2022]
Abstract
OBJECTIVE Excessive bleeding is an important complication of radical cystectomy. We aimed to assess whether preoperative administration of fibrinogen decreases perioperative bleeding and improves the outcome of radical cystectomy. DESIGN Double-blinded randomized trial with two parallel arms. SETTING The study was conducted in the department of surgery at a teaching hospital affiliated with a University of Medical Sciences. PATIENTS In total, 70 men undergoing radical cystectomy were randomized to fibrinogen (n = 35) and placebo-control groups. Mean (SD) age was 64.7 (7.4) years. INTERVENTIONS The intervention group received 2 g fibrinogen concentrate diluted in 100 ml distilled water, and the control group received 100 ml normal saline; both intravenously 15 ̶ 30 min before the start of the surgery. OUTCOME MEASURES The primary outcome was the amount of perioperative blood loss. The secondary outcomes were hemodynamic features and vital signs. MAIN RESULTS Fibrinogen significantly decreased the volume of blood loss (p < 0.001) and the total number of transfused packed-cell units per group (38 vs. 115 units); and compensated the decrease of HCO3 (p = 0.030), the mean arterial pressure (p < 0.001), hemoglobin O2 saturation (p = 0.001), heart rate (p < 0.001), and temperature (p < 0.001) throughout the surgery compared with the placebo. Patients in the fibrinogen group had shorter Intensive Care Unit (p = 0.001) and hospital (p < 0.001) stay. We did not find any adverse reaction in our patients receiving fibrinogen concentrate. CONCLUSION Fibrinogen concentrate reduces perioperative bleeding and the need for blood transfusion in radical cystectomy. It improves the outcomes of the surgery and decreases patients' length of stay in the healthcare system following radical cystectomy. REGISTRATION Iranian Registry of Clinical Trials (IRCT) http://www.irct.ir/, reference number: IRCT20191013045091N1. ETHICS CODE Shahid Beheshti University of Medical Sciences, reference number: IR.SBMU.RETECH.REC.1398.033.
Collapse
|
6
|
Dunn LK, Chen CJ, Taylor DG, Esfahani K, Brenner B, Luo C, Buell TJ, Spangler SN, Buchholz AL, Smith JS, Shaffrey CI, Nemergut EC, Durieux ME, Naik BI. Postoperative Low-Dose Tranexamic Acid After Major Spine Surgery: A Matched Cohort Analysis. Neurospine 2020; 17:888-895. [PMID: 33401867 PMCID: PMC7788407 DOI: 10.14245/ns.2040114.057] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2020] [Revised: 04/22/2020] [Accepted: 05/08/2020] [Indexed: 01/04/2023] Open
Abstract
OBJECTIVE This was a retrospective, cohort study investigating the efficacy and safety of continuous low-dose postoperative tranexamic acid (PTXA) on drain output and transfusion requirements following adult spinal deformity surgery. METHODS One hundred forty-seven patients undergoing posterior instrumented thoracolumbar fusion of ≥ 3 vertebral levels at a single institution who received low-dose PTXA infusion (0.5-1 mg/kg/hr) for 24 hours were compared to 292 control patients who did not receive PTXA. The cohorts were propensity matched based on age, sex, American Society of Anesthesiologist physical status classification, body mass index, number of surgical levels, revision surgery, operative duration, and total intraoperative TXA dose (n = 106 in each group). Primary outcome was 72-hour postoperative drain output. Secondary outcomes were number of allogeneic blood transfusions. RESULTS There was no significant difference in postoperative drain output in the PTXA group compared to control (660 ± 420 mL vs. 710 ± 490 mL, p = 0.46). The PTXA group received significantly more crystalloid (6,100 ± 3,100 mL vs. 4,600 ± 2,400 mL, p < 0.001) and red blood cell transfusions postoperatively (median [interquartile range]: 1 [0-2] units vs. 0 [0-1] units; incidence rate ratio [95% confidence interval], 1.6 [1.2-2.2]; p = 0.001). Rates of adverse events were comparable between groups. CONCLUSION Continuous low-dose PTXA infusion was not associated with reduced drain output after spinal deformity surgery. No difference in thromboembolic incidence was observed. A prospective dose escalation study is warranted to investigate the efficacy of higher dose PTXA.
Collapse
Affiliation(s)
- Lauren K. Dunn
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Ching-Jen Chen
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Davis G. Taylor
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Kamilla Esfahani
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Brian Brenner
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Charles Luo
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Thomas J. Buell
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Sarah N. Spangler
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Avery L. Buchholz
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Justin S. Smith
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Christopher I. Shaffrey
- Departments of Neurosurgery and Orthopaedic Surgery, Duke University Medical Center, Durham, NC, USA
| | - Edward C. Nemergut
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Marcel E. Durieux
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| | - Bhiken I. Naik
- Department of Anesthesiology, University of Virginia Health Science Center, Charlottesville, VA, USA
- Department of Neurological Surgery, University of Virginia Health Science Center, Charlottesville, VA, USA
| |
Collapse
|