RESUMEN
Background: Patients with atrial fibrillation (AF) remain at increased risk of thromboembolism despite apparent maintenance of sinus rhythm with the cause often attributed to periods of asymptomatic AF. Atrial mechanical discordance, with the body of the left atrium (LA) in sinus rhythm and the left atrial appendage (LAA) in AF may also be a contributor. Objectives: The purpose of this study was to assess the frequency of electrocardiogram (ECG) rhythm and LAA and/right atrial appendage (RAA) Doppler ejection phenotype (transesophageal echocardiography [TEE]) discordance in patients undergoing cardiac surgery. Methods: A total of 124 patients undergoing coronary artery bypass graft (CABG), CABG and valve surgery, or isolated valve repair or replacement (valve ± CABG) were prospectively studied. Intraoperative surface ECG rhythm strip and TEE were performed before cardiopulmonary bypass. The ECG and TEE LAA/RAA Doppler spectrum were independently classified as sinus or AF. Results: Of 107 patients (age 65 ± 12 years; 31% female; 65% CABG, 31% valve ± CABG) without a history of AF, 39 (36%) had ECG and LAA and/or RAA discordance (ECG/LAA Doppler discordance, n = 12 [11%]; ECG/RAA Doppler discordance, n = 35 [33%]). There was no significant difference between concordant and discordant groups with regard to age, gender, history of hypertension, diabetes, heart failure, or stroke (all P > 0.05). Conclusions: A large minority of patients without a history of AF undergoing cardiac surgery have ECG/atrial appendage Doppler discordance, a setting that may promote thromboembolism in non-anticoagulated patients. Clinical parameters do not identify patients at increased risk for discordance.
RESUMEN
PURPOSE: Small left atrial (LA) volume was recently reported to be one of the best predictors of acute pulmonary embolism (PE)-related adverse events (AE). There is currently no data available regarding the impact that body surface area (BSA)-indexing of atrial measurements has on the association with PE-related adverse events. Our aim is to assess the impact of indexing atrial measurements to BSA on the association between computed tomography (CT) atrial measurements and AE. MATERIALS AND METHODS: Retrospective study (IRB: 2015P000425). A database of hospitalized patients with acute PE diagnosed on CT pulmonary angiography (CTPA) between May 2007 and December 2014 was reviewed. Right and left atrial volume, largest axial area, and axial diameters were measured. Patients undergo both echocardiographies (from which the BSA was extracted) and CTPAs within 48 hours of the procedure. The patient's body weight was measured during each admission. LA measurements were correlated to AE (defined as the need for advanced therapy or PE-related mortality at 30 days) before and after indexing for BSA. The area under the ROC curve was calculated to determine the predictive value of the atrial measurements in predicting AE. RESULTS: The study included 490 acute PE patients; 62 (12.7%) had AE. There was a significant association of reduced BSA-indexed and non-indexed LA volume (both <0.001), area (<0.001 and 0.001, respectively), and short-axis diameters (both <0.001), and their respective RA/LA ratios (all <0.001) with AE. The AUC values were similar for BSA-indexed and non-indexed LA volume, diameters, and area with LA volume measurements being the best predictor of adverse outcomes (BSA-indexed AUC=0.68 and non-indexed AUC=0.66), followed by non-indexed LA short-axis diameter (indexed AUC=0.65, non-indexed AUC=0.64), and LA area (indexed AUC=0.64, non-indexed AUC=0.63). CONCLUSION: Adjusting for BSA does not substantially affect the predictive ability of atrial measurements on 30-day PE-related adverse events, and therefore, this adjustment is not necessary in clinical practice. While LA volume is the better predictor of AE, LA short-axis diameter has a similar predictive value and is more practical to perform clinically.