Risk of Atrial Fibrillation and Associated Outcomes After Allogeneic Hematopoietic Stem Cell Transplantation

Get Permission

In a single-institution study reported in the Journal of Clinical Oncology, Chang et al found that atrial fibrillation occurs in a substantial proportion of patients who have undergone allogeneic hematopoietic cell transplantation (HCT) and is associated with poor outcomes.

Study Details

The retrospective cohort study involved 509 consecutive patients undergoing allogeneic HCT at City of Hope Comprehensive Cancer Center between January 2014 and December 2016. The population included in the analysis consisted of 487 patients after exclusion of patients aged < 18 years (n = 65), those with prior history of atrial fibrillation (n = 34), and those undergoing a second HCT (n = 13).

Multivariate analysis was adjusted for patient demographics, comorbidities, pre-HCT electrocardiogram (ECG) parameters, and HCT-related exposures with P < .1 on univariate analysis. A nested case-control study was performed to assess associations between pre-HCT echocardiograms and atrial fibrillation events. Control patients were matched with cases for age at HCT, sex, and length of follow-up.

Photo credit: Getty

Incidence and Risk of Atrial Fibrillation

Median follow-up was 3.1 years (range = 0.0–6.0 years), with 81% of the cohort being followed through December 2019 for surviving patients or death. Median age at HCT was 52.4 years (range = 18.1–78.6 years). The majority of patients were male (57.7%) and White (48.7%). The most common diagnoses were acute myeloid leukemia (42.3%), acute lymphoblastic leukemia (25.7%), and myelodysplastic syndromes or myeloproliferative neoplasms (21.6%).

A total of 50 patients developed atrial fibrillation; 437 did not develop atrial fibrillation. The cumulative incidence of atrial fibrillation at 1 and 5 years was 6.8% and 10.6%. Median time to atrial fibrillation was 117.5 days (range = 4.0–1,405.0 days). Among the 50 patients with atrial fibrillation, 3 (6%) had a lone atrial fibrillation and 47 (94%) had paroxysmal or persistent atrial fibrillation.

On univariate analysis, factors significantly associated with increased risk of atrial fibrillation were older age (≥ 50 years) at HCT, male sex, diabetes, dyslipidemia, pre-HCT prolonged QTc interval, human leukocyte antigen (HLA)-unrelated donor, and nonmyeloablative conditioning.

On multivariate analysis (n = 468 with available electrocardiograms), factors significantly associated with increased atrial fibrillation risk were older (≥ 50 years) age (hazard ratio [HR] = 2.76, 95% confidence interval [CI] = 1.37–5.58), HLA-unrelated donor (HR = 2.20, 95% CI = 1.18–4.12), dyslipidemia (HR = 2.40, 95% CI = 1.23–4.68), and pre-HCT prolonged QTc interval (HR = 2.55, 95% CI = 1.38–4.72).


  • The cumulative incidence of atrial fibrillation at 5 years was 10.6%.
  • Atrial fibrillation was associated with increased risk of all-cause and nonrelapse mortality.

A total of 39 patients with atrial fibrillation (78%) and 389 without atrial fibrillation (89%) had available pre-HCT echocardiograms. There were no significant differences in clinical or treatment characteristics between patients with vs without pre-HCT echocardiograms. In the case-control analysis among 39 patients with atrial fibrillation and 97 without atrial fibrillation, there was no difference in median left-ventricular ejection fraction between groups (61% vs 60%, P = .582). However, atrial fibrillation cases had significantly lower median left-atrial ejection fraction (39.7% vs 53.4%, P < .001) and left-atrial reservoir function (31.1% vs 39.1%, P = .001) and a higher prevalence of abnormal tricuspid regurgitant jet velocity (30.8% vs 8.2%, P = .006).  

In multivariate analysis in the case-control population, significantly increased risk for atrial fibrillation was associated with abnormal left-atrial ejection fraction (< 45%; odds ratio [OR] = 12.7, P < .001), low left-atrial reservoir function (< 39%; OR = 3.8, P = .010), and elevated tricuspid regurgitant jet velocity (> 2.8m/s; OR = 4.2, P = .023). The odds ratio for presence of two or more abnormal measures was 18.0 (P < .001).


Among the 50 patients with atrial fibrillation, 5 (10.0%) developed stroke during 34.8 person-years of follow-up after atrial fibrillation onset, yielding an incidence rate of 143 per 1,000 person-years. Overall survival in the entire cohort at 1 and 5 years was 75.4% and 55.8%, with the most common cause of death being relapse or disease progression (15.8% of cohort). Overall survival was 20.0% among the 50 patients with atrial fibrillation vs 62.0% among the 437 without atrial fibrillation. Mortality due to relapse or disease progression was 12.0% vs 16.2%. Nonrelapse mortality was higher among atrial fibrillation patients, including death due to multiorgan failure (20.0% vs 3.7%), graft-vs-host disease (12.0% vs 5.0%), cardiovascular causes (14.0% vs 1.8%), sepsis (14.0% vs 3.2%), and pneumonia/respiratory failure (8.0% vs 3.7%).      

In adjusted analysis, atrial fibrillation was associated with significantly increased risk of all-cause mortality (HR = 12.76, 95% CI = 8.76–18.57) and nonrelapse mortality (HR = 15.78, 95% CI = 8.70–28.62), with no significant difference in risk of relapse-related mortality (HR = 1.08, 95% CI = 0.43–2.68).

The investigators concluded, “The burden of atrial fibrillation after allogeneic HCT population is substantial, and the development of atrial fibrillation is associated with poor survival. We identified important associations between patient demographics, pre-HCT cardiac parameters, HCT-related exposures, and risk of atrial fibrillation, setting the stage for targeted prevention strategies during and after HCT.”

Saro Armenian, DO, MPH, of the City of Hope Comprehensive Cancer Center, is the corresponding author for the Journal of Clinical Oncology article.

Disclosure: The study was supported by grants from the Lymphoma/Leukemia Society and National Institutes of Health/National Heart, Lung, and Blood Institute. For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.