TY - JOUR
T1 - Cardiovascular Risk Stratification of Patients Undergoing Hematopoietic Stem Cell Transplantation
T2 - The CARE-BMT Risk Score
AU - Vasbinder, Alexi
AU - Catalan, Tonimarie
AU - Anderson, Elizabeth
AU - Chu, Catherine
AU - Kotzin, Megan
AU - Murphy, Danielle
AU - Cheplowitz, Halle
AU - Diaz, Kristen Machado
AU - Bitterman, Brayden
AU - Pizzo, Ian
AU - Huang, Yiyuan
AU - Xie, Jeffrey
AU - Hoeger, Christopher W.
AU - Kaakati, Rayan
AU - Berlin, Hanna P.
AU - Shadid, Husam
AU - Perry, Daniel
AU - Pan, Michael
AU - Takiar, Radhika
AU - Padalia, Kishan
AU - Mills, Jamie
AU - Meloche, Chelsea
AU - Bardwell, Alina
AU - Rochlen, Matthew
AU - Blakely, Pennelope
AU - Leja, Monika
AU - Banerjee, Mousumi
AU - Riwes, Mary
AU - Magenau, John
AU - Anand, Sarah
AU - Ghosh, Monalisa
AU - Pawarode, Attaphol
AU - Yanik, Gregory
AU - Nathan, Sunita
AU - Maciejewski, John
AU - Okwuosa, Tochukwu
AU - Hayek, Salim S.
N1 - Publisher Copyright:
© 2023 The Authors.
PY - 2024/1/2
Y1 - 2024/1/2
N2 - BACKGROUND: Evidence guiding the pre-hematopoietic stem cell transplantation (HSCT) cardiovascular evaluation is limited. We sought to derive and validate a pre-HSCT score for the cardiovascular risk stratification of HSCT candidates. METHODS AND RESULTS: We leveraged the CARE-BMT (Cardiovascular Registry in Bone Marrow Transplantation) study, a contemporary multicenter observational study of adult patients who underwent autologous or allogeneic HSCT between 2008 and 2019 (N=2435; mean age at transplant of 55 years; 4.9% Black). We identified the subset of variables most predictive of post-HSCT cardiovascular events, defined as a composite of cardiovascular death, myocardial infarction, heart failure, stroke, atrial fibrillation or flutter, and sustained ventricular tachycardia. We then developed a point-based risk score using the hazard ratios obtained from Cox proportional hazards modeling. The score was externally validated in a separate cohort of 919 HSCT recipients (mean age at transplant 54 years; 20.4% Black). The risk score included age, transplant type, race, coronary artery disease, heart failure, peripheral artery disease, creatinine, triglycerides, and prior anthracycline dose. Risk scores were grouped as low-, intermediate-, and high-risk, with the 5-year cumulative incidence of cardiovascular events being 4.0%, 10.3%, and 22.4%, respectively. The area under the receiver operating curves for predicting cardiovascular events at 100 days, 5 and 10 years post-HSCT were 0.65 (95% CI, 0.59–0.70), 0.73 (95% CI, 0.69–0.76), and 0.76 (95% CI, 0.69–0.81), respectively. The model performed equally well in autologous and allogeneic recipients, as well as in the validation cohort. CONCLUSIONS: The CARE-BMT risk score is easy to calculate and could help guide referrals of high-risk HSCT recipients to cardiovascular specialists before transplant and guide long-term monitoring.
AB - BACKGROUND: Evidence guiding the pre-hematopoietic stem cell transplantation (HSCT) cardiovascular evaluation is limited. We sought to derive and validate a pre-HSCT score for the cardiovascular risk stratification of HSCT candidates. METHODS AND RESULTS: We leveraged the CARE-BMT (Cardiovascular Registry in Bone Marrow Transplantation) study, a contemporary multicenter observational study of adult patients who underwent autologous or allogeneic HSCT between 2008 and 2019 (N=2435; mean age at transplant of 55 years; 4.9% Black). We identified the subset of variables most predictive of post-HSCT cardiovascular events, defined as a composite of cardiovascular death, myocardial infarction, heart failure, stroke, atrial fibrillation or flutter, and sustained ventricular tachycardia. We then developed a point-based risk score using the hazard ratios obtained from Cox proportional hazards modeling. The score was externally validated in a separate cohort of 919 HSCT recipients (mean age at transplant 54 years; 20.4% Black). The risk score included age, transplant type, race, coronary artery disease, heart failure, peripheral artery disease, creatinine, triglycerides, and prior anthracycline dose. Risk scores were grouped as low-, intermediate-, and high-risk, with the 5-year cumulative incidence of cardiovascular events being 4.0%, 10.3%, and 22.4%, respectively. The area under the receiver operating curves for predicting cardiovascular events at 100 days, 5 and 10 years post-HSCT were 0.65 (95% CI, 0.59–0.70), 0.73 (95% CI, 0.69–0.76), and 0.76 (95% CI, 0.69–0.81), respectively. The model performed equally well in autologous and allogeneic recipients, as well as in the validation cohort. CONCLUSIONS: The CARE-BMT risk score is easy to calculate and could help guide referrals of high-risk HSCT recipients to cardiovascular specialists before transplant and guide long-term monitoring.
KW - atrial fibrillation
KW - bone marrow transplant
KW - cardiovascular disease
KW - heart failure
KW - hematopoietic stem cell transplant
KW - random forest
KW - risk score
UR - http://www.scopus.com/inward/record.url?scp=85181585980&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85181585980&partnerID=8YFLogxK
U2 - 10.1161/JAHA.123.033599
DO - 10.1161/JAHA.123.033599
M3 - Article
C2 - 38158222
AN - SCOPUS:85181585980
SN - 2047-9980
VL - 13
JO - Journal of the American Heart Association
JF - Journal of the American Heart Association
IS - 1
M1 - e033599
ER -