Predicting transfusions in cardiac surgery

The easier, the better: The Transfusion Risk and Clinical Knowledge score

M. Ranucci, S. Castelvecchio, A. Frigiola, S. Scolletta, P. Giomarelli, B. Biagioli

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

Background and Objectives Allogeneic blood products transfusions are associated with an increased morbidity and mortality risk in cardiac surgery. At present, a few transfusion risk scores have been proposed for cardiac surgery patients. The present study is aimed to develop and validate a risk score based on adequate statistical analyses joint with a clinical selection of a limited (five) number of preoperative predictors. Materials and Methods The development series was composed of 8989 consecutive adult patients undergone cardiac surgery. Independent predictors of allogeneic blood transfusions were identified. Subsequently, five predictors were extracted as the most clinically relevant based on the judgement of 30 clinicians dealing with transfusions in cardiac surgery. A predictive score was developed and externally validated on a series of 2371 patients operated in another institution. The score was compared to the other existing scores. Results The following predictors constituted the Transfusion Risk and Clinical Knowledge score: age > 67 years; weight <60 kg for females and <85 kg for males preoperative haematocrit; gender - female; and complex surgery. At the external validation, this score demonstrated an acceptable predictive power (area under the curve 0·71) and a good calibration at the Hosmer-Lemeshow test. When compared to the other three existing risk scores, the Transfusion Risk and Clinical Knowledge score had comparable or better predictive power and calibration. Conclusion A simple risk model based on five predictors only has a similar or better accuracy and calibration in predicting the transfusion rate in cardiac surgery than more complex models.

Original languageEnglish
Pages (from-to)324-332
Number of pages9
JournalVox Sanguinis
Volume96
Issue number4
DOIs
Publication statusPublished - May 2009

Fingerprint

Thoracic Surgery
Calibration
Blood Transfusion
Hematocrit
Area Under Curve
Joints
Morbidity
Weights and Measures
Mortality

Keywords

  • RBC transfusions
  • Transfusion practices (adults)

ASJC Scopus subject areas

  • Hematology

Cite this

Predicting transfusions in cardiac surgery : The easier, the better: The Transfusion Risk and Clinical Knowledge score. / Ranucci, M.; Castelvecchio, S.; Frigiola, A.; Scolletta, S.; Giomarelli, P.; Biagioli, B.

In: Vox Sanguinis, Vol. 96, No. 4, 05.2009, p. 324-332.

Research output: Contribution to journalArticle

@article{7136cc94aaf04a5aaa49f38e206eee0b,
title = "Predicting transfusions in cardiac surgery: The easier, the better: The Transfusion Risk and Clinical Knowledge score",
abstract = "Background and Objectives Allogeneic blood products transfusions are associated with an increased morbidity and mortality risk in cardiac surgery. At present, a few transfusion risk scores have been proposed for cardiac surgery patients. The present study is aimed to develop and validate a risk score based on adequate statistical analyses joint with a clinical selection of a limited (five) number of preoperative predictors. Materials and Methods The development series was composed of 8989 consecutive adult patients undergone cardiac surgery. Independent predictors of allogeneic blood transfusions were identified. Subsequently, five predictors were extracted as the most clinically relevant based on the judgement of 30 clinicians dealing with transfusions in cardiac surgery. A predictive score was developed and externally validated on a series of 2371 patients operated in another institution. The score was compared to the other existing scores. Results The following predictors constituted the Transfusion Risk and Clinical Knowledge score: age > 67 years; weight <60 kg for females and <85 kg for males preoperative haematocrit; gender - female; and complex surgery. At the external validation, this score demonstrated an acceptable predictive power (area under the curve 0·71) and a good calibration at the Hosmer-Lemeshow test. When compared to the other three existing risk scores, the Transfusion Risk and Clinical Knowledge score had comparable or better predictive power and calibration. Conclusion A simple risk model based on five predictors only has a similar or better accuracy and calibration in predicting the transfusion rate in cardiac surgery than more complex models.",
keywords = "RBC transfusions, Transfusion practices (adults)",
author = "M. Ranucci and S. Castelvecchio and A. Frigiola and S. Scolletta and P. Giomarelli and B. Biagioli",
year = "2009",
month = "5",
doi = "10.1111/j.1423-0410.2009.01160.x",
language = "English",
volume = "96",
pages = "324--332",
journal = "Vox Sanguinis",
issn = "0042-9007",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - Predicting transfusions in cardiac surgery

T2 - The easier, the better: The Transfusion Risk and Clinical Knowledge score

AU - Ranucci, M.

AU - Castelvecchio, S.

AU - Frigiola, A.

AU - Scolletta, S.

AU - Giomarelli, P.

AU - Biagioli, B.

PY - 2009/5

Y1 - 2009/5

N2 - Background and Objectives Allogeneic blood products transfusions are associated with an increased morbidity and mortality risk in cardiac surgery. At present, a few transfusion risk scores have been proposed for cardiac surgery patients. The present study is aimed to develop and validate a risk score based on adequate statistical analyses joint with a clinical selection of a limited (five) number of preoperative predictors. Materials and Methods The development series was composed of 8989 consecutive adult patients undergone cardiac surgery. Independent predictors of allogeneic blood transfusions were identified. Subsequently, five predictors were extracted as the most clinically relevant based on the judgement of 30 clinicians dealing with transfusions in cardiac surgery. A predictive score was developed and externally validated on a series of 2371 patients operated in another institution. The score was compared to the other existing scores. Results The following predictors constituted the Transfusion Risk and Clinical Knowledge score: age > 67 years; weight <60 kg for females and <85 kg for males preoperative haematocrit; gender - female; and complex surgery. At the external validation, this score demonstrated an acceptable predictive power (area under the curve 0·71) and a good calibration at the Hosmer-Lemeshow test. When compared to the other three existing risk scores, the Transfusion Risk and Clinical Knowledge score had comparable or better predictive power and calibration. Conclusion A simple risk model based on five predictors only has a similar or better accuracy and calibration in predicting the transfusion rate in cardiac surgery than more complex models.

AB - Background and Objectives Allogeneic blood products transfusions are associated with an increased morbidity and mortality risk in cardiac surgery. At present, a few transfusion risk scores have been proposed for cardiac surgery patients. The present study is aimed to develop and validate a risk score based on adequate statistical analyses joint with a clinical selection of a limited (five) number of preoperative predictors. Materials and Methods The development series was composed of 8989 consecutive adult patients undergone cardiac surgery. Independent predictors of allogeneic blood transfusions were identified. Subsequently, five predictors were extracted as the most clinically relevant based on the judgement of 30 clinicians dealing with transfusions in cardiac surgery. A predictive score was developed and externally validated on a series of 2371 patients operated in another institution. The score was compared to the other existing scores. Results The following predictors constituted the Transfusion Risk and Clinical Knowledge score: age > 67 years; weight <60 kg for females and <85 kg for males preoperative haematocrit; gender - female; and complex surgery. At the external validation, this score demonstrated an acceptable predictive power (area under the curve 0·71) and a good calibration at the Hosmer-Lemeshow test. When compared to the other three existing risk scores, the Transfusion Risk and Clinical Knowledge score had comparable or better predictive power and calibration. Conclusion A simple risk model based on five predictors only has a similar or better accuracy and calibration in predicting the transfusion rate in cardiac surgery than more complex models.

KW - RBC transfusions

KW - Transfusion practices (adults)

UR - http://www.scopus.com/inward/record.url?scp=65649127180&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65649127180&partnerID=8YFLogxK

U2 - 10.1111/j.1423-0410.2009.01160.x

DO - 10.1111/j.1423-0410.2009.01160.x

M3 - Article

VL - 96

SP - 324

EP - 332

JO - Vox Sanguinis

JF - Vox Sanguinis

SN - 0042-9007

IS - 4

ER -