Derivation and validation of a nomogram model for pulmonary thromboembolism in patients undergoing lung cancer surgery
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Derivation and validation of a nomogram model for pulmonary thromboembolism in patients undergoing lung cancer surgery. / Li, Yuping; Shen, Lei; Ding, Junrong; Xie, Dong; Yang, Jian; Zhao, Yanfeng; Carretta, Angelo; Petersen, René Horsleben; Gilbert, Sebastien; Hida, Yasuhiro; Bölükbas, Servet; Fernando, Hiran C; Jiang, Gening; Zhu, Yuming.
In: Translational Lung Cancer Research, Vol. 10, No. 4, 2021, p. 1829-1840.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Derivation and validation of a nomogram model for pulmonary thromboembolism in patients undergoing lung cancer surgery
AU - Li, Yuping
AU - Shen, Lei
AU - Ding, Junrong
AU - Xie, Dong
AU - Yang, Jian
AU - Zhao, Yanfeng
AU - Carretta, Angelo
AU - Petersen, René Horsleben
AU - Gilbert, Sebastien
AU - Hida, Yasuhiro
AU - Bölükbas, Servet
AU - Fernando, Hiran C
AU - Jiang, Gening
AU - Zhu, Yuming
N1 - 2021 Translational Lung Cancer Research. All rights reserved.
PY - 2021
Y1 - 2021
N2 - Background: A specific risk-stratification tool is needed to facilitate safe and cost-effective approaches to the prophylaxis of acute pulmonary thromboembolism (PTE) in lung cancer surgery patients. This study aimed to develop and validate a simple nomogram model for the prediction of PTE after lung cancer surgery using readily obtainable clinical characteristics.Methods: A total of 14,427 consecutive adult patients who underwent lung cancer surgery between January 2015 and July 2018 in our institution were retrospectively reviewed. Included in the cohort were 136 patients who developed PTE and 544 non-PTE patients. The patients were randomly divided into the derivation group (70%, 95 PTE patients and 380 non-PTE patients) and the validation group (30%, 41 PTE patients and 164 non-PTE patients). A nomogram model was developed based on the results of multivariate logistic analysis in the derivation group. The cut-off values were defined using Youden's index. The prognostic accuracy was measured by area under the curve (AUC) values.Results: In the derivation group, multivariate logistic analysis was carried out to evaluate the risk score. The risk assessment model contained five variables: age [95% confidence interval (CI): 1.008-1.083, P=0.016], body mass index (95% CI: 1.077-1.319, P=0.001), operation time (95% CI: 1.002-1.014, P=0.008), the serum level of cancer antigen 15-3 (CA15-3) before surgery (95% CI: 1.019-1.111, P=0.005), and the abnormal results of compression venous ultrasonography before surgery (95% CI: 2.819-18.838, P<0.001). All of them were independent risk factors of PTE. To simplify the risk assessment model, a nomogram model was established, which showed a good predictive performance in the derivation group (AUC 0.792, 95% CI: 0.734-0.853) and in the validation group (AUC 0.813, 95% CI: 0.737-0.890).Conclusions: A high-performance nomogram was established on the risk factors for PTE in patients undergoing lung cancer surgery. The nomogram could be used to provide an individual risk assessment and guide prophylaxis decisions for patients. Further external validation of the model is needed in lung cancer surgery patients in other clinical centers.
AB - Background: A specific risk-stratification tool is needed to facilitate safe and cost-effective approaches to the prophylaxis of acute pulmonary thromboembolism (PTE) in lung cancer surgery patients. This study aimed to develop and validate a simple nomogram model for the prediction of PTE after lung cancer surgery using readily obtainable clinical characteristics.Methods: A total of 14,427 consecutive adult patients who underwent lung cancer surgery between January 2015 and July 2018 in our institution were retrospectively reviewed. Included in the cohort were 136 patients who developed PTE and 544 non-PTE patients. The patients were randomly divided into the derivation group (70%, 95 PTE patients and 380 non-PTE patients) and the validation group (30%, 41 PTE patients and 164 non-PTE patients). A nomogram model was developed based on the results of multivariate logistic analysis in the derivation group. The cut-off values were defined using Youden's index. The prognostic accuracy was measured by area under the curve (AUC) values.Results: In the derivation group, multivariate logistic analysis was carried out to evaluate the risk score. The risk assessment model contained five variables: age [95% confidence interval (CI): 1.008-1.083, P=0.016], body mass index (95% CI: 1.077-1.319, P=0.001), operation time (95% CI: 1.002-1.014, P=0.008), the serum level of cancer antigen 15-3 (CA15-3) before surgery (95% CI: 1.019-1.111, P=0.005), and the abnormal results of compression venous ultrasonography before surgery (95% CI: 2.819-18.838, P<0.001). All of them were independent risk factors of PTE. To simplify the risk assessment model, a nomogram model was established, which showed a good predictive performance in the derivation group (AUC 0.792, 95% CI: 0.734-0.853) and in the validation group (AUC 0.813, 95% CI: 0.737-0.890).Conclusions: A high-performance nomogram was established on the risk factors for PTE in patients undergoing lung cancer surgery. The nomogram could be used to provide an individual risk assessment and guide prophylaxis decisions for patients. Further external validation of the model is needed in lung cancer surgery patients in other clinical centers.
U2 - 10.21037/tlcr-21-109
DO - 10.21037/tlcr-21-109
M3 - Journal article
C2 - 34012796
VL - 10
SP - 1829
EP - 1840
JO - Translational Lung Cancer Research
JF - Translational Lung Cancer Research
SN - 2218-6751
IS - 4
ER -
ID: 276375744