1
|
Chen X, Pan J, Li Y, Tang R. Application of machine learning model in predicting the likelihood of blood transfusion after hip fracture surgery. Aging Clin Exp Res 2023; 35:2643-2656. [PMID: 37733228 DOI: 10.1007/s40520-023-02550-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/05/2023] [Accepted: 08/24/2023] [Indexed: 09/22/2023]
Abstract
OBJECTIVE Anemia is one of the common adverse reactions after hip fracture surgery. The traditional method to solve anemia is allogeneic transfusion. However, the transfusion may lead to some complications such as septicemia and fever. So far, few studies have reported roles of machine learning in predicting whether blood transfusion is needed or not after hip fracture surgery. Therefore, the purpose of this study is to develop machine learning models to predict the likelihood of postoperative blood transfusion in patients undergoing hip fracture surgery. METHODS This study enrolled 1355 patients who underwent hip fracture surgery at the Affiliated Hospital of Qingdao University from January 2016 to December 2021. Among all patients, 210 cases received postoperative blood transfusion. All patients were randomly divided into a training group and a testing group at a ratio of 7:3. In the training group, univariate and multivariate logistic regression analyses were used to determine independent risk factors for the postoperative transfusion. Then, based on these independent risk factors, tenfold cross-validation method was utilized to develop five machine learning models, including logistic, multilayer perceptron (MLP), extreme gradient boosting (XGBoost), random forest (RF), and support vector machine (SVM). The receiver operating characteristic (ROC) curve, area under ROC curve (AUC), and Matthews correlation coefficient (MCC) were generated to evaluate the performance of the models. Calibration plot and decision curve analysis (DCA) were used to test the performance, stability, and clinical applicability of the models. The models were validated using the testing group; and the ROC curve, MCC, calibration plot, and DCA curves were also generated to validate the performance, stability, and clinical applicability of the models. To further verify the robustness of the model, we randomly grabbed 70% of the samples in the testing set, performed 1000 iterations, and calculated the AUC and confidence interval of the five models. Finally, we used SHapley Additive exPlanations (SHAP) to explain these models. RESULTS Multivariate logistic regression analysis showed that there were 8 independent risk factors, including age, blood transfusion history, albumin (ALB), globulin (GLO), total bilirubin (TBIL), indirect bilirubin (IBIL), hemoglobin (HB), and blood loss > 200 ml. We finally selected five independent risk factors including HB, GLO, age, IBIL, and blood loss > 200 ml. Based on these five independent risk factors, we generated six characteristic variables, namely HB, HB × HB, HB × blood loss, GLO × HB, age, age × IBIL, and established five machine learning models using a tenfold cross-validation method. In the training group, the AUC values of logistic, RF, MLP, SVM, and XGB were 0.9320, 0.8911, 0.9327, 0.9225, and 0.8825, respectively, and the average AUC was 0.9122 ± 0.0212. The MCC values were 0.65, 0.77, 0.65, 0.66, and 0.68, respectively, and the calibration plot and DCA performed well. In the testing group the AUC values of logistic, RF, MLP, SVM, and XGB were 0.8483, 0.7978, 0.8576, 0.8598, and 0.8216, respectively. The average AUC was 0.8370 ± 0.0238, and the MCC values were 0.41, 0.35, 0.40, 0.41, and 0.41, respectively. The calibration plot and DCA in the testing group also showed good performance. The AUC values and confidence intervals of the 1000-iteration model were: logistic (AUC, min confidence interval [CI]-max confidence interval [CI] 0.848, 0.804-0.903), RF (AUC, minCI-maxCI 0.797, 0.734-0.857), MLP (AUC, minCI-maxCI 0.858, 0.812-0.902), SVM (AUC, minCI-maxCI 0.859, 0.819-0.910), and XGB (AUC, minCI-maxCI 0.821, 0.764-0.894). The model performed well. Finally, according to SHAP, among all five models, HB played the most important role in model prediction and interpretation. CONCLUSION The five models we developed all performed well in predicting the likelihood of blood transfusion after hip fracture surgery. Therefore, we believed that the prediction model based on machine learning had great application prospects in clinical practice, which could help clinicians better predict the risk of blood transfusion after hip fracture surgery.
Collapse
Affiliation(s)
- Xiao Chen
- Department of Orthopaedics, Suzhou Hospital of Anhui Medical University, Suzhou, 234000, Anhui, China
| | - Junpeng Pan
- Department of Spinal Surgery, The Affiliated Hospital of Qingdao University, Qingdao, 266000, Shandong, China
| | - Yi Li
- Department of Orthopaedics, Suzhou Hospital of Anhui Medical University, Suzhou, 234000, Anhui, China
| | - Ruixin Tang
- Department of Orthopaedics, Suzhou Hospital of Anhui Medical University, Suzhou, 234000, Anhui, China.
| |
Collapse
|
2
|
Effectiveness of Perioperative Comprehensive Evaluation of Hip Fracture in the Elderly. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2022; 2022:4124354. [PMID: 36035830 PMCID: PMC9410974 DOI: 10.1155/2022/4124354] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Received: 05/26/2022] [Revised: 07/13/2022] [Accepted: 07/16/2022] [Indexed: 12/31/2022]
Abstract
Objective The objective is to observe the effect of Comprehensive Geriatric Assessment (CGA) in the perioperative period of hip fracture. Methods From October 2018 to October 2021, 155 patients over the age of 65 diagnosed with hip fracture and treated with surgery at the Department of Trauma Orthopaedics of General Hospital of Ningxia Medical University were randomly divided into two groups using a prospective research method. A total of 70 cases in the CGA group received a perioperative comprehensive assessment of the geriatric, and 85 cases in the control group received routine medical consultation. Results Elderly patients with hip fractures have a high comorbidity index. Patients with abnormal daily activity before injury accounted for 55%, the abnormal rate of nutrition was 58.1%, the abnormal rate of cognition, anxiety, and depression was 81.8%, and 77.3% of the patients were in a weak state. There was no significant difference in age, gender, ASA grade, fracture type, and operation mode between the two groups, but there were significant differences in operation rate at 48 h (χ2 = 22.153; P ≤ 0.001), preoperative waiting time (Z = −6.387; P ≤ 0.001), total hospital stay (Z = −11.756; P ≤ 0.001), and incidence of postoperative delirium (χ2 = 23.897; P ≤ 0.001). Conclusions The implementation of CGA shortened the preoperative waiting time and total hospital stay, increased the 48 h operation rate, and reduced the incidence of postoperative delirium.
Collapse
|
3
|
Li Y, Tung KK, Cho YC, Lin SY, Lee CH, Chen CH. Improved outcomes and reduced medical costs through multidisciplinary co-management protocol for geriatric proximal femur fractures: a one-year retrospective study. BMC Geriatr 2022; 22:318. [PMID: 35410173 PMCID: PMC8996524 DOI: 10.1186/s12877-022-03014-6] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/19/2021] [Accepted: 03/24/2022] [Indexed: 01/13/2023] Open
Abstract
BACKGROUND To manage the rapidly growing incidence of, and related medical burden resulting from hip fractures in older adults in an aging society, studies involving orthogeriatric co-management treatment models have reported improved outcomes, including reduced medical costs. The treatment gap for osteoporosis was however seldom emphasized in the published treatment protocols. Aiming to improve the existing orthogeriatric protocol, we have established a patient-centered protocol for elderly patient hip fractures, which simultaneously focuses on fracture care and anti-osteoporosis agent prescription in regarding to healthcare quality and medical expense. METHODS This was a retrospective study comparing patients who enrolled in the multidisciplinary co-managed protocol for geriatric hip fractures and those who did not. The inclusion criteria for this study were: (a) single-sided hip fractures treated from 1 to 2018 to 30 June 2020, (b) patients who were 60-years or older (c) trauma treated within 3 days from time of injury, and (d) minimal follow-up period of 12 months after surgery. RESULTS From 1 to 2018 to 30 June 2020, 578 patients were included (267 patients in the protocol group vs. 331 patients in the conventional group). The protocol group was associated with significantly reduced lengths of hospital stay (p = 0.041), medical expenditures (p = 0.006), and mortality (p = 0.029) during their acute in-hospital admission period. Early osteoporosis diagnosis and anti-osteoporosis agent prescription were achieved in the protocol group, with a significantly wider coverage for BMD assessment (p < 0.001) and prescriptions for anti-osteoporosis medication (p < 0.001). Yet, there was no significant decline in the one-year refracture rate in the protocol group. CONCLUSIONS The implementation of a multidisciplinary co-managed care protocol for geriatric proximal femur fractures successfully improved patient outcomes with significantly reduced lengths of stay, medical expenditures, and mortality during the acute in-hospital admission period. The high prescription rate of anti-osteoporosis medication after hip fractures in the protocol group was not associated with a significantly lower re-fracture rate in the 12-month follow-up. However, the association between early anti-osteoporosis agent prescription and reduced long-term medical expenses in this group of patients has provided a direction for future research.
Collapse
Affiliation(s)
- Yang Li
- Department of Orthopedics, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Kuan-Kai Tung
- Department of Orthopedics, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Yi-Cheng Cho
- Department of Orthopedics, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Shih-Yi Lin
- School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan.,Center for Geriatrics and Gerontology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan.,Division of Endocrinology and Metabolism, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
| | - Cheng-Hung Lee
- Department of Orthopedics, Taichung Veterans General Hospital, Taichung, Taiwan.,School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan
| | - Chih-Hui Chen
- School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan. .,Department of Orthopedics, Changhua Christian Hospital, Changhua, Taiwan. .,, No. 135, Nanxiao St, Changhua County, 50006, Changhua City, Taiwan.
| |
Collapse
|
4
|
Levi AR, Coste M, Warshowsky E, Shah NV, Suneja N, Schwartz JM, Roudnitsky V. Cracking the Hip: Does Protocol Matter? A Retrospective Cohort Study Investigating the Effect of Protocol Implementation. Geriatr Orthop Surg Rehabil 2022; 13:21514593221076614. [PMID: 35242395 PMCID: PMC8886300 DOI: 10.1177/21514593221076614] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/01/2021] [Revised: 01/04/2022] [Accepted: 01/17/2022] [Indexed: 11/20/2022] Open
Abstract
Introduction Approximately 300 000 hip fractures occur annually in the USA in patients
>65 years old. Early intervention is key in reducing morbidity and
mortality. Our institution implemented a collaborative hip fracture
protocol, streamlining existing processes to reduce time to OR (TTO) and
hospital length of stay (LOS). Our aim was to determine if this protocol
improved these outcomes. Study Design We conducted a retrospective cohort study using our level-1 trauma center’s
trauma registry, comparing outcomes for patients >60 years old with
isolated hip fractures pre-and post-hip protocol implementation in May 2018.
Our primary outcomes were TTO and in-hospital mortality. Secondary outcomes
included LOS and postoperative complications. Univariate analysis was done
using chi-square and T-test. Results We identified 176 patients with isolated hip fractures: 69 post- and 107
pre-protocol. Comparing post- to pre-protocol, TTO decreased by 18hrs (39 vs
57h; P = .013) and patients had fewer postoperative
complications (9 vs 23%; P = .016) despite post-protocol
patients being more likely to have diabetes (42 vs 27%, P
< .05), elevated BMI (22 vs 25; P < .001), and to be
current smokers (9 vs 2%; P < .05). LOS and in-hospital
mortality also decreased (11 vs 20d; P = .312, 4.3 vs 7.5%;
P = .402). Post-protocol patients were more likely to
go to the OR within 24hrs of presentation (39 vs 16%; P
< .001) and to go straight from ED to OR (32 vs 4%; P
< .001). Conclusion TTO, LOS, and postoperative complications for isolated hip fracture patients
were lower post-protocol. Though not all statistically significant, this
trend indicates that the protocol was helpful in improving hip fracture
outcomes but may require further improvement and institution-wide
education.
Collapse
Affiliation(s)
- Amelia R Levi
- Department of Surgery, SUNY Downstate Health & Sciences University, Kings County NYC Health & Hospitals, MedStar Georgetown University Hospital, Washington, DC, USA
| | - Marine Coste
- Department of Surgery, SUNY Downstate Health & Sciences University, Kings County NYC Health & Hospitals, Brooklyn, NY, USA
| | - Ethan Warshowsky
- Department of Surgery, SUNY Downstate Health & Sciences University, Kings County NYC Health & Hospitals, Brooklyn, NY, USA
| | - Neil V Shah
- Department of Orthopaedic Surgery, SUNY Downstate Health & Sciences University, Kings County NYC Health and Hospitals, Brooklyn, NY, USA
| | - Nishant Suneja
- Department of Orthopaedic Surgery, Kings County NYC Health and Hospitals, Brooklyn, NY, USA
| | - Jeffrey M Schwartz
- Department of Orthopaedic Surgery, Kings County NYC Health and Hospitals, Brooklyn, NY, USA
| | - Valery Roudnitsky
- Department of Surgery, SUNY Downstate Health & Sciences University, Kings County NYC Health & Hospitals, Brooklyn, NY, USA
| |
Collapse
|
5
|
Fan J, Lv Y, Xu X, Zhou F, Zhang Z, Tian Y, Ji H, Guo Y, Yang Z, Hou G. The Efficacy of Multidisciplinary Team Co-Management Program for Elderly Patients With Intertrochanteric Fractures: A Retrospective Study. Front Surg 2022; 8:816763. [PMID: 35284470 PMCID: PMC8907576 DOI: 10.3389/fsurg.2021.816763] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/17/2021] [Accepted: 12/28/2021] [Indexed: 11/13/2022] Open
Abstract
BackgroundIntertrochanteric fractures increased quickly in past decades owing to the increasing number of aging population. Recently, geriatric co-management was rapidly emerging as a favored clinical care model for older patients with hip fractures. The purpose of this study was to assess the efficacy of a multidisciplinary team (MDT) co-management program in elderly patients with intertrochanteric fractures.MethodsIn this retrospective study, patients were divided into MDT group and traditional orthopedic care (TOC) group according to the healthcare model applied. 249 patients were included in the TOC group from January 2014 to December 2016 and 241 patients were included in the MDT group from January 2017 to December 2019. Baseline data, peri-operative data, and postoperative complications were collected and analyzed using SPSS 21.0.ResultsNo significant differences were observed between the two groups in terms of patient baseline characteristics. Patients in the MDT group had significantly lower time from admission to surgery and length-of-stay (LOS) compared with those in the TOC group. Furthermore, the proportion of patients receiving surgery within 24 h (61.4 vs. 34.9%, p < 0.001) and 48 h (80.9 vs. 63.5%, p < 0.001) after admission to the ward was significantly higher in the MDT group compared with those in the TOC group. In addition, patients in the MDT group had significantly lower proportion of postoperative complications (25.3 vs. 44.2%, p < 0.001), deep vein thrombosis (7.9 vs. 12.9%, p = 0.049), pneumonia (3.8 vs. 8.0%, p = 0.045) and delirium (4.1 vs. 9.2%, p = 0.025) compared with those in the TOC group. However, no significant changes were found for in-hospital and 30-day mortality.ConclusionThe MDT co-management could significantly shorten the time from admission to surgery, LOS, and reduce the postoperative complications for elderly patients with intertrochanteric fractures. Further research was needed to evaluate the impact of this model on patient health outcomes.
Collapse
Affiliation(s)
- Jixing Fan
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Yang Lv
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Xiangyu Xu
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Fang Zhou
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
- *Correspondence: Fang Zhou
| | - Zhishan Zhang
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Yun Tian
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Hongquan Ji
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Yan Guo
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Zhongwei Yang
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| | - Guojin Hou
- Department of Orthopedics, Peking University Third Hospital, Beijing, China
- Engineering Research Center of Bone and Joint Precision Medicine, Ministry of Education, Beijing, China
| |
Collapse
|
6
|
Subcuticular Barbed Suture and Skin Glue Wound Closure Decreases Reoperation and Length of Stay in Geriatric Hip Fractures When Compared With Staples. JOURNAL OF THE AMERICAN ACADEMY OF ORTHOPAEDIC SURGEONS GLOBAL RESEARCH AND REVIEWS 2021; 5:01979360-202110000-00005. [PMID: 34605793 PMCID: PMC8492365 DOI: 10.5435/jaaosglobal-d-21-00205] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Received: 08/12/2021] [Accepted: 08/12/2021] [Indexed: 11/18/2022]
Abstract
INTRODUCTION Patients with geriatric hip fracture are notoriously frail and at risk for complications. Persistent postoperative wound drainage can lead to prolonged hospital stay, increased risk for infection, and need for revision surgery. The purpose of this study was to determine the effect of wound closure technique, barbed monofilament subcuticular suture and skin glue versus staples on rates of intervention for wound drainage and length of hospital stay after geriatric hip fracture fixation. METHODS A retrospective review of isolated hip fractures in patients older than 60 years at a single institution over a 3-year period was done. Hip fractures included femoral neck, intertrochanteric, and subtrochanteric femur fractures treated with internal fixation or arthroplasty. Skin closure technique, at the discretion of the operating surgeon, included either barbed subcuticular monofilament suture and skin glue or staples. Charts and radiographs were reviewed to determine patient characteristics, Charlson Comorbidity Index, type of wound closure, length of stay, and interventions for persistent wound drainage. RESULTS There were 175 patients in the barbed suture and skin glue group and 211 patients in the staples group. The barbed suture group had an average postsurgical length of stay of 5.0 days which was significantly lower than the staples group (7.0 days, P < 0.00001). In the staples group, 17 patients (8%) required incisional negative pressure wound therapy due to wound drainage with five patients (2.4%) returning to the operating room secondary to persistent wound drainage. No patients were observed in the barbed suture group that required intervention for wound drainage. DISCUSSION Barbed suture and skin glue closure is associated with markedly shorter hospital stay and fewer interventions for wound drainage when compared with staples after surgical treatment of geriatric hip fractures.
Collapse
|