Engraftment Day Electrolytes and Hydration and the Risk of Atrial Fibrillation Early after Hematopoietic Stem Cell Transplant

Background Atrial fibrillation (AF) in hematopoietic stem cell transplant (HSCT) recipients is associated with significant morbidity. We have previously reported that obstructive sleep apnea (OSA), older age and presence of dilated left atrium (LA) are significant predictors of AF early post HSCT. In the current study, we sought to evaluate if TSH pre HSCT and serum electrolytes and eGFR at engraftment in patients undergoing HSCT are associated with increased risk of AF. Methods This is a single center retrospective study, involving 748 consecutive patients undergoing autologous and allogenic HSCT from 2012 to 2022. Patients’ charts were reviewed to acquire clinical information (age, gender, HTN, body mass index [BMI], Obstructive sleep apnea [OSA], TSH, and LA volume index that was obtained from pre-HSCT echocardiogram, in addition to potassium [K], magnesium [Mg], and GFR at time of engraftment [ENG]). Results For the 748 patients, the median age at HSCT was 61. Majority of patients were male (57%). Most common diagnoses were Myeloma (42.5%), acute leukemia (19.5%), lymphoma (19.5%), myelodysplastic syndromes and myeloproliferative neoplasms (11%). A total of 116 (15.5%) patients developed AF early post-HSCT. AF patients’ mean LA measurement was 37.3ml/m2, whereas non-AF patients’ mean measure was 30.6ml/m2 (P-value: 0.002). For age at HSCT, AF patients’ mean was 62.4 years old and non-AF patients’ mean was 57.4 years old (P-value: <0.001). OSA diagnosis, 39 (29.5%) of AF patients had OSA and 67 (11.3%) of non-AF patients had OSA (P-value: <0.001). At time of ENG, AF patients had higher mean K of 4.1 (P-value: 0.001) and lower mean GFR of 81.1 (P-value<0.001) versus non-AF patients with mean K of 3.7 and GFR of 94.3. In the multivariable regression analysis, larger LA size (P-value: 0.016), older age at HSCT (P-value: 0.001), diagnosis of OSA (P-value: <0.001), and higher K (P-value: 0.010) and lower GFR (P-value: 0.007) at ENG were significant predictors of AF. Male gender, HTN, BMI, use of ATG or Melphalan based chemotherapies, pre-HSCT TSH, and Mg at ENG were not predictive of AF post-HSCT (table 1). Conclusion In our single center retrospective study, we found that compared to patients who did not develop AF early post-HSCT, the AF patients, in addition to having dilated LA, being older and more OSA, surprisingly they had higher K and lower GFR at day of ENG. A limitation of our study is the diagnosis of OSA, which sometimes was not supported with a sleep study or diagnosis came from an external practitioner. Another limitation of our study would be the fluctuations in electrolytes around time of engraftment and overhydration due to IV fluids during the inpatient stay. Further research is needed to develop a risk calculator to identify high risk patients and study the effects of prophylactic therapies on the incidence of post-HSCT AF.