Analysis of a large data set to identify predictors of blood transfusion in primary total hip and knee arthroplasty

Huang, Zeyu; Huang, Cheng; Xie, Jinwei; Ma, Jun; Cao, Guorui; Huang, Qiang; Shen, Bin; Kraus, Virginia B.; Pei, Fuxing

doi:10.1111/trf.14783

Cited by 47 publications

(54 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The mean age of this cohort was 54.8 years, in contrast to previously reported 64.7–71.0 years in western countries. However, data of 7477 patients from 26 university teaching hospitals in China suggested that the mean age of Chinese THA patients was 57.6 ± 18.2 years. Besides, Chan et al .…”

Section: Discussioncontrasting

confidence: 84%

“…High prevalence of osteonecrosis of femoral head (ONFH) and developmental dysplasia of the hip (DDH) in contrast to very low prevalence of primary hip osteoarthritis have been observed in the Chinese population. Besides, due to different races, the Chinese THA patients have much lower age and BMI than those reported abroad. As a result, we considered that the predictors for LOS in western literatures may be nonequivalent for those of Chinese patients.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Limited Influence of Comorbidities on Length of Stay after Total Hip Arthroplasty: Experience of Enhanced Recovery after Surgery

Ding

Liang

et al. 2019

Orthopaedic Surgery

View full text Add to dashboard Cite

Objectives To identify predictors of length of stay (LOS) after total hip arthroplasty (THA) in an enhanced recovery after surgery (ERAS) program and evaluate the safety and cost‐efficiency of the ERAS program with reduced LOS for unselected patients in a Chinese population. Methods A total of 311 consecutive, unselected patients undergoing primary THA at a single institution were retrospectively reviewed and divided into two groups: LOS ≤ 3 and LOS > 3 group. All patients were managed with the same ERAS protocol and went back home after discharge. Multivariate logistic regression analysis was used to determine independent risk factors for LOS > 3. Harris Hip Score at 90‐day follow‐up, 90‐day readmission rate, and hospitalization costs were compared between two groups. Results Multivariate regression analysis identified female gender (odds ratio [OR] = 2.623), living alone (OR = 4.127), and primary osteoarthritis of hip (OR = 3.565) to be correlated with LOS > 3. Preoperative hemoglobin (HB), postoperative HB, drain use, blood transfusion, diabetes, respiratory disease, osteoporosis, number of comorbidities, and CCI score showed no significant influence on LOS after adjusting for other risk factors in the multivariate model. Harris Hip Score and readmission rate at 90‐day follow‐up showed no significant differences between two groups. Patients in LOS > 3 group had approximately 3948.6 Chinese yuan higher hospital costs. Conclusion Female gender, living alone, and primary osteoarthritis of hip were identified as independent risk factors for prolonged LOS. The experience from our institution suggested aggressive management of comorbidities in the ERAS program can minimize the influence of comorbidities on LOS. The safety, efficiency, and costs‐saving benefits of the ERAS program with reduced LOS for unselected patients was confirmed in this study.

show abstract

Section: Discussioncontrasting

confidence: 84%

Section: Introductionmentioning

confidence: 99%

Limited Influence of Comorbidities on Length of Stay after Total Hip Arthroplasty: Experience of Enhanced Recovery after Surgery

Ding

Liang

et al. 2019

Orthopaedic Surgery

View full text Add to dashboard Cite

show abstract

“…Using modern machine learning tools, we could reliably predict which patients will be in need for transfusion of RBCs with a manageable number of features provided at hospital admission. In contrast to other studies, we did not restrict prediction to one patient group, like others have done 18,19 but provided a model that can be used over a broad range of indications. However, although this prediction has proven to be very reliable for classification of "transfusion" vs "no transfusion," we found out for our database that the total number of RBC transfusions per patient and the occurrence of massive transfusion in a specific patient cannot be predicted reliably, a phenomenon that has been described in liver transplantations by other groups.…”

Section: Discussionmentioning

confidence: 99%

Machine learning–based prediction of transfusion

et al. 2020

View full text Add to dashboard Cite

Background The ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning–based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital. Study Design and Methods This was a retrospective, observational study in three adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures for the classification tasks were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the four machine learning algorithms used: neural networks (NNs), logistic regression (LR), random forests (RFs), and gradient boosting (GB) trees. Results Using our four predictive models, transfusion of at least 1 unit of RBCs could be predicted rather accurately (sensitivity for NN, LR, RF, and GB: 0.898, 0.894, 0.584, and 0.872, respectively; specificity: 0.958, 0.966, 0.964, 0.965). Using the four methods for prediction of massive transfusion was less successful (sensitivity for NN, LR, RF, and GB: 0.780, 0.721, 0.002, and 0.797, respectively; specificity: 0.994, 0.995, 0.993, 0.995). As a consequence, prediction of the total number of packed RBCs transfused was also rather inaccurate. Conclusion This study demonstrates that the necessity for intrahospital transfusion can be forecasted reliably, however the amount of RBC units transfused during a hospital stay is more difficult to predict.

show abstract

“…A previous study [13] based on a nationwide inpatient sample database reported that TKA and THA were among the top 10 most-frequent procedures requiring transfusion in the USA. Furthermore, blood transfusion was related to numerous serious complications such as viral infection, anaphylactic and hemolytic reactions, longer PLOS, deep vein thromboembolism, high hospitalization costs, and increased in-hospital mortality [14–16]. Allogeneic blood reserves are often insufficient worldwide; consequently, reducing the transfusion rate is essential for patients undergoing TKA or THA.…”

Section: Discussionmentioning

confidence: 99%

Closed suction drainage following routine primary total joint arthroplasty is associated with a higher transfusion rate and longer postoperative length of stay: a retrospective cohort study

Xie

Lei

et al. 2019

J Orthop Surg Res

Self Cite

View full text Add to dashboard Cite

Background In an enhanced recovery after surgery program, not placing a closed suction drain following routine primary total joint arthroplasty (TJA) is becoming more acceptable. However, the influence of drain use on transfusion rate and postoperative length of stay (PLOS) in TJA remains controversial. Therefore, we aimed to compare drain use with no drain in routine primary TJA to determine the differences in transfusion rate and PLOS. Methods We analyzed the data from 12,992 patients undergoing primary unilateral TJA: 6325 total knee arthroplasties (TKA) and 6667 total hip arthroplasties (THA). Patients were divided into two groups according to whether they received a drain postoperatively following TKA and THA. We extracted information for transfusion and PLOS from patients’ electronic health records and analyzed the data by logistic and linear regression analyses. Results The transfusion rate and PLOS were 15.07% and 7.75 ± 3.61 days, respectively, in the drain group and 6.72% and 6.54 ± 3.32 days, respectively, in the no-drain group following TKA. The transfusion rate and PLOS were 20.53% and 7.00 ± 3.35 days, respectively, in the drain group and 13.57% and 6.07 ± 3.06 days, respectively, in the no-drain group following THA. After adjusting for the following variables: age, gender, body mass index, orthopedic diagnoses, hypertension, type 2 diabetes, coronary heart disease, chronic obstructive pulmonary disease, preoperative hemoglobin, albumin, analgesic use, anesthesia, American Society of Anesthesiologists class, tranexamic acid use, intraoperative bleeding, operative time, and tourniquet use (for TKA), drain use correlated significantly with a higher transfusion rate (risk ratio = 2.812, 95% confidence interval (CI) 2.224–3.554, P < 0.001 for TKA and risk ratio = 1.872, 95% CI 1.588–2.207, P < 0.001 for THA) and a longer PLOS (partial regression coefficient ( B ) = 1.099, 95% CI 0.879–1.318, P < 0.001, standard regression coefficient ( B ′) = 0.139 for TKA; B = 0.973, 95% CI 0.695–1.051, P < 0.001, and B ′ = 0.115 for THA). Two groups showed no significant difference in wound complications. Conclusions Our findings indicated that drain use was associated with a higher transfusion rate and a longer PLOS in patients undergoing routine primary TJA. The routine use of postoperative drainage is not recommended in primary unilateral TJA.

show abstract

Analysis of a large data set to identify predictors of blood transfusion in primary total hip and knee arthroplasty

Cited by 47 publications

References 49 publications

Limited Influence of Comorbidities on Length of Stay after Total Hip Arthroplasty: Experience of Enhanced Recovery after Surgery

Limited Influence of Comorbidities on Length of Stay after Total Hip Arthroplasty: Experience of Enhanced Recovery after Surgery

Machine learning–based prediction of transfusion

Closed suction drainage following routine primary total joint arthroplasty is associated with a higher transfusion rate and longer postoperative length of stay: a retrospective cohort study

Contact Info

Product

Resources

About