OUP user menu

Left atrial volumetry from routine diagnostic work up prior to pulmonary vein ablation is a good predictor of freedom from atrial fibrillation

Christian Sohns , Jan M. Sohns , Dirk Vollmann , Lars Lüthje , Leonard Bergau , Marc Dorenkamp , Paul A. Zwaka , Gerd Hasenfuß , Joachim Lotz , Markus Zabel
DOI: http://dx.doi.org/10.1093/ehjci/jet017 684-691 First published online: 22 February 2013


Aims This study aimed to identify whether left atrial (LA) volume assessed by multidetector computed tomography (MDCT) is related to the long-term success of pulmonary vein ablation (PVA). MDCT is used to guide PVA for the treatment of atrial fibrillation (AF). MDCT permits accurate sizing of LA dimensions.

Methods and results We analysed data from 368 ablation procedures of 279 consecutive patients referred for PVA due to drug-refractory symptomatic AF (age 62 ± 10; 58% men; 71% paroxysmal AF). Prior to the procedure, all patients underwent ECG-gated 64-MDCT scan for assessment of LA and PV anatomy, LA thrombus evaluation, LA volume estimation, and electroanatomical mapping integration. Within a mean follow-up of 356 ± 128 days, 64% of the patients maintained sinus rhythm after the initial ablation, and 84% when including repeat PVA. LA diameter (P = 0.004), LA volume (P = 0.002), and type of AF (P = 0.001) were independent predictors of AF recurrence in univariate analysis. There was a relatively low correlation between the echocardiographic LA diameter and LA volume from MDCT (P = 0.01, r = 0.5). In multivariate analysis, paroxysmal AF (P < 0.006) and LA volume below the median value of 106 mL (P = 0.042) were significantly associated with the success of PVA, whereas LA diameter was not (P = 0.245). Analysing receiver-operator characteristics, the area under the curve for LA volume was 0.73 (P = 0.001) compared with 0.60 (P = 0.09) for LA diameter from echocardiography.

Conclusion LA volume assessed by MDCT is a better predictor of AF recurrence after PVA than echocardiograpic LA diameter and can be derived from the pre-procedural imaging data set.

  • Atrial fibrillation
  • Pulmonary vein ablation
  • Multidetector computed tomography
  • Left atrium


Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in the adult and is associated with decreased quality of life as well as increased morbidity and mortality.1,2 Percutaneous radiofrequency (RF) catheter ablation (RFCA) of AF with pulmonary vein ablation (PVA) has evolved over the last decade and is currently recommended for symptomatic patients refractory to antiarrhythmic drug therapy.3 Imaging of the left atrium (LA) and related intrathoracic structures before PVA is valuable for anatomical guidance of the procedure.4,5 Multidetector computed tomography (MDCT) allows three-dimensional visualization of patient-specific cardiac anatomy including the LA, LA appendage (LAA), as well as the number and anatomy of the pulmonary veins (PV). Furthermore, pre-processed three-dimensional MDCT data sets can be integrated into advanced electroanatomical mapping systems to guide the operator during the mapping and ablation procedure, and to avoid complications.6 The recurrence rates of AF after PVA vary between studies; reasons for this include different ablation protocols, divergent modes of follow-up and changing definitions of AF recurrence, and success following the PVA procedure.7,8 LA size is a well-known risk factor for AF recurrence after PVA.7 However, echocardiographic LA diameter may not represent the true extent of LA enlargement as measured by LA volumetry.9 This suggests that the prognostic content of LA dimensional parameters for the probability of AF recurrence after PVA may be underestimated. Routine diagnostic work up prior to PVA and using ECG-gated 64-MDCT permits highly accurate assessment of LA anatomy and dimensions. Our aim was therefore to prospectively study if pre-procedural quantification of LA volume is better associated with the outcome following PVA.


Patient selection

A total of 279 consecutive patients with drug-refractory, symptomatic paroxysmal or persistent AF were scheduled to undergo PVA between October 2007 and April 2011 at the Heart Center of Georg-August-University Göttingen, Germany. All clinical, imaging, and procedural data were prospectively recorded. Written informed consent was obtained from each patient prior to the procedures. Paroxysmal AF was defined as self-terminating episodes of AF lasting <7 days. Persistent AF was defined as AF sustained >7 days, and/or requiring electrical or pharmacological cardioversion.1,10 All patients underwent transthoracic echocardiography (TTE), transoesophageal echocardiography (TEE) to rule out intracardiac thrombus, and contrast-enhanced MDCT for anatomical evaluation of the LA and PV. Exclusion criteria were hyperthyroidism, LA thrombus, decompensated heart failure, stroke, myocardial infarction or gastrointestinal bleeding within 4 weeks prior to the intervention, and life-expectancy <6 months.

Multidetector computed tomography

All patients underwent ECG-gated 64-slice MDCT (VCT LightSpeed, GE Healthcare, Milwaukee, WI, USA) within 24–48 h prior to PVA using our local LA protocol.11 Briefly, heart rate was kept below 70 bpm using β-adrenergic blocking agents if necessary. A spiral scan using retrospective gating was performed during a single breath-hold to examine the heart from the supraaortic region to the upper abdomen. In all cases, 80 mL of intravenous contrast agent (Imeron 350, Bracco Imaging, Konstanz, Germany) was injected. Semi-automatic bolus detection in the ascending aorta was used for optimal contrast timing. Imaging parameters included gantry rotation time of 350 ms, detector collimation of 64 × 0.625 mm, and a tube voltage of 120 kV. Semi-automatic dose reduction schemes were used. The radiation exposures range between 6.24 and 10.31 mSv. An ECG-gated half-scan algorithm was applied to reconstruct the data into axial images with a slice thickness of 0.625 mm. Ten phases within the cardiac cycle were reconstructed at 10% RR interval and the end-systolic phase of the LA selected as either 70 or 80% of RR length. MDCT images were analysed offline by experienced independent readers (at least one radiologist and one cardiologist) on a standard workstation with a dedicated cardiac imaging software package (VolumeShare 2, GE Healthcare). For each patient, anatomy of the LA, the PVs, and LA volume was identified. As previously described, the two-dimensional LA area was manually traced on each MDCT slice from the LA roof to the level of the mitral annulus. The PVs were cut at the PV ostia and the LAA was excluded at its base.12 The LA volume was automatically calculated and reported. For the PVA procedure, 3D MDCT images of the LA and the PV were reconstructed on a separate workstation and integrated with electroanatomical mapping (CARTO Merge, Biosense Webster, Diamond Bar, CA, USA).11


LA diameter, left ventricular (LV) ejection fraction, and LV end-diastolic diameter were assessed by TTE. The TTE images were obtained from parasternal long- and short-axis views, apical four-chamber, two-chamber, and long-axis views. Chambers were quantified according to the standards described by the American Society of Echocardiography (ASE).13 The LA diameter was measured ∼1 cm above the mitral annulus in the two- and four-chamber views. In addition, to recognize thrombus formation in the LA or LAA appendage, all patients underwent TEE within 24 h of the PVA procedure. The TEE was performed using a GE Vivid E9 ultrasound system (GE Ultrasound, Horten, Norway) with a 5.0 MHz multiplane probe. Cine loops of the LAA were acquired during a continuous sweep from 0° to 180°. The presence of LA thrombus was considered if there was an intracavitary echogenic mass that could be distinguished from the surrounding tissue in more than one imaging plane.11

Electrophysiological study

For the electrophysiological procedure, all catheters were advanced via the femoral vein. A 6 F steerable decapolar catheter (Bard Dynamic Tip, Bard Inc., Lowell, MA, USA) was positioned in the coronary sinus. After fluoroscopically guided transseptal puncture, a 3.5 mm open-irrigated, magnetic mapping and ablation catheter (Navistar Thermocool RMT, Biosense Webster) was advanced through an SL1 sheath (St Jude Medical, Inc., St Paul, MN, USA) into the LA. Circumferential PV ablation was performed using a 3D-mapping system (CARTO Merge, Biosense Webster) in conjunction with the integrated MDCT image of the LA and real-time fluoroscopy. The Niobe II magnetic navigation system (Stereotaxis) and a joystick-controlled motor drive (Cardiodrive, Stereotaxis) were utilized for remote magnetic navigation (RMN) of the ablation catheter in patients with RMN-guided ablation, whereas the catheter was guided manually in patients with conventional ablation. The RMN system has been previously described.14,15 The RF generator (Stockert, Biosense Webster) was set to temperature-controlled RF delivery with a target temperature of 45°C and a nominal power limit of 40 W (flow 30 mL/min). At the posterior LA wall, the output was limited to 30 W (flow 17 mL/ min). RF current was applied for 30–60 s until local electrogram amplitude was reduced by 80%. The endpoint of the ablation procedure was the electrical isolation of all PVs defined as bidirectional conduction block. This was verified by careful and repeated mapping for residual potentials around the entire circumference of the PV ostia, and pacing from multiple sites within the circumferential line.15


Daily 12-lead surface ECGs and telemetry until discharge were used to confirm sinus rhythm. Additional antiarrhythmic drug therapy was individually prescribed for the first 3–6 months. Oral anticoagulation was started the day after the procedure with a target INR of 2.0–3.0. Bridging with unfractionated or low molecular weight heparin was initiated 6–12 h after the procedure. After hospital discharge, all patients were scheduled in our outpatient clinic 3, 6, 9, and 12 months after PVA. Upon every follow-up visit, patients were asked for symptoms of AF, documented arrhythmia recurrences, and current medication. Moreover, ambulatory Holter monitoring was performed at each visit for 96 h to detect symptomatic or asymptomatic AF or atrial tachycardia (AT) recurrences. All patients were advised to present themselves immediately in the case of symptoms suggestive of AT/AF recurrence in order to obtain necessary treatment and ECG documentation. An AT/AF episode lasting longer than 30 s outside a blanking period of 3 months after the index procedure was considered as recurrent AT/AF.

Statistical analysis

Statistical analysis was performed using SPSS for Windows (Version 17.0, SPSS Inc., Chicago, IL, USA). Continuous variables are expressed as mean ± standard deviation or as median with inter-quartile range if appropriate. Normally distributed data were compared using the independent Student's t-test. A P-value of <0.05 is considered as statistically significant.

A Kaplan–Meier analysis of the median value of a continuous parameter was used as the standard cut-off value. Freedom of AF/AT recurrence for the dichotomized patient groups was compared using a log-rank test. Multivariate logistic or Cox regression analysis was performed to assess those factors that achieved significance in univariate analysis. Receiver operating characteristic (ROC) analysis was performed to assess and compare the predictive value of parameters for ablation success. Area under the ROC curve (AUC) as well as the asymptotic significance were calculated.


We enrolled 279 consecutive patients, aged 62 ± 10 years; 58% were men. Totally, we analysed data from 368 ablation procedures. Two hundred and forty-seven PVA were (67%) RMN-guided and 121 procedures (33%) were performed conventional. The clinical baseline characteristics of the entire study population are presented in Table 1. The majority of patients had paroxysmal AF (71%) and were anticoagulated at the time of the pre-procedural investigations (78%). All patients underwent TTE, TEE, and MDCT prior to PVA. Before the ablation procedure, patients had been treated with amiodarone (48%), dronedarone (29%), and flecainide (74%) (Table 1).

View this table:
Table 1

Baseline patient characteristics

Age (years)62 ± 10
Male sex213 (58%)
Paroxysmal AF262 (71%)
AF duration (years)6.3 ± 5.9
Prior amiodarone175 (48%)
Prior dronedarone38 (29%)
Prior flecainide272 (74%)
Coronary artery disease52 (14%)
Dilated cardiomyopathy18 (5%)
Valvular heart disease (≥2°)59 (16%)
Hypertension281 (74%)
Diabetes mellitus61 (17%)
Serum creatinine (mg/dL)0.92 ± 0.25
Prior stroke/TIA33 (9%)
LA diameter (mm)43 ± 8
LA volume (mL)111 ± 29
PAPsys (mmHg)34 ± 9
LV ejection fraction (%)55 ± 8.9
LV ejection fraction ≤40%40 (11%)
AF at the time of MDCT140 (38%)
Age ≥75 years26 (7%)
  • AF, atrial fibrillation; MDCT, Multidetector computed tomography.

The mean follow-up per patient was 365 ± 128 days. Considering only the initial procedure, 64% of the patients were free of AT/AF recurrences (Figure 2A). Including repeat PVA procedures (1.7 ± 0.6 procedures), 84% maintained sinus rhythm during the follow-up period (Figure 2B). Antiarrhythmic drugs were prescribed to 19% of the patients without symptomatic AF recurrence. Of those patients, 10% were on amiodarone.

The Kaplan–Meier analysis showed that the probability of AT/AF recurrence after initial successful PVA was significantly higher in patients with non-paroxysmal AT/AF (P = 0.001), and LA volume above the median of 106 mL (P = 0.024). The LA diameter above the median of 45 mm measured by TTE showed a strong tendency to predict freedom from AT/AF. Nonetheless, it was not a statistically significant parameter for the success of the ablation procedure (P = 0.063, Figure 2C). Within the data set, a relatively low correlation between the LA diameter measured by TTE and the LA volume assessed by MDCT was found by linear (P = 0.01, r = 0.51; Figure 3) and by exponential regression (P = 0.01, r = 0.52; Figure 3).

Other clinical baseline parameters such as age (P = 0.095), gender (P = 0.452), body mass index (P = 0.584), arterial hypertension (P = 0.054), diabetes mellitus (P = 0.255), septum thickness (MDCT P = 0.31, TTE P = 0.07), and LV ejection fraction (P = 0.213) did not correlate with the success of the procedure (Table 2). This was also true for prior antiarrhythmic drug treatment with amiodarone (P = 0.329), dronedarone (P = 0.365), or flecainide (P = 0.125) (Table 2). In the multivariate Cox regression analysis, paroxysmal AF (P = 0.006) and LA volume <106 mL (P = 0.042) remained significantly associated with the success of the procedure, whereas LA diameter was not (P = 0.245). The LA dimensions were also categorized according to the ASE classification, as reference range, mildly enlarged, moderately enlarged, and severely enlarged (Table 3).16

View this table:
Table 2

Measured parameters in patients with and without recurrence of AF/AT

VariableTotalAT/AF recurrenceNo recurrenceP-value
Univariate analysisMultivariate analysis
Age (years)62 ± 1061 ± 562 ± 100.095NS
Gender (male)213 (58%)75 (57%)138 (58%)0.452NS
BMI (kg/m²)31 ± 730 ± 431 ± 50.584NS
Paroxsymal/persistent AF262 (71%)/105 (29%)81 (62%)/50 (38%)181 (76%)/55 (24%)0.001a0.006a
Diabetes mellitus6119420.255NS
LA volume (MDCT) (mL)111 ± 29117 ± 32107 ± 260.002a0.042a
Septum thickness (MDCT) (mm)10,04 ± 0,510 ± 0,2410,06 ± 0,650.318NS
LA diameter (TTE) (mm)43 ± 844 ± 542 ± 80.004aNS
LV end-diastolic diameter (TTE) (mm)50 ± 550 ± 349 ± 60.052NS
LV ejection fraction55 ± 8.953 ± 1.857 ± 1,20.213NS
Septum thickness (TTE) (mm)10.06 ± 0.3210.1 ± 0.3310.04 ± 0.310.072NS
Prior amiodarone175 (48%)65 (49%)110 (46%)0.329NS
Prior dronedarone112 (30%)38 (29%)74 (31%)0.365NS
Prior flecainide272 (74%)102 (78%)170 (72%)0.125NS
Mean follow-up (days)335 ± 63343 ± 63330 ± 630.058NS
  • BMI, body mass index; MDCT, multidetector computed tomography; TTE, transthoracic echocardiography; LV, left ventricle.

  • aStatistically significant; ns, statistically not significant in multivariate analysis.

View this table:
Table 3

Reference limits and partition values for LA dimensions/volumes according to the ASE classification16

Reference rangeMildly abnormalModerately abnormalSeverely abnormalReference rangeMildly abnormalModerately abnormalSeverely abnormal
LA diameter (mm)27–3839–4243–46≥4730–4041–4647–52≥52
LA volume (mL)22–5253–6263–72≥7318–5859–6869–78≥79

LA volume measurement by MDCT categorized reliable more patients in the severely enlarged group (177 patients) than the LA diameter investigated by TTE (68 patients). With regard to predict the recurrence of AT/AF after successful PVA, the strongest predictor of long-term PVA failure was a severely enlarged LA volume measured by MDCT. In this context, the AUC from ROC analysis for the LA volume was 0.73 (P < 0.001) and 0.6 (P = 0.09) for the LA diameter.

Comparing the prognostic accuracy of the two imaging modalities by means of ROC curve analysis, LA volume exhibited an AUC of 0.73 (P < 0.001). In contrast, for echocardiographic LA diameter, an AUC of 0.60 was calculated which was not significant (P = 0.09). The Kaplan–Meier analysis showed a significantly lower AT/AF free probability in patients below the median LA volume of 106 mL than in those above (P = 0.024, Figure 2D).


The main finding of this prospective study is that the LA volume assessed by ECG-gated MDCT prior to circumferential PVA reliably predicts the success and recurrence of AT/AF after RFCA ablation, and does so better than echocardiographic LA diameter. To our knowledge, this is one of the first prospective studies that aimed to evaluate LA dimensions and its prediction of freedom from AF/AT after PVA by different imaging modalities in a large patient cohort including 368 ablation procedures. Our analysis shows that independent risk markers for the recurrence of AT/AF after successful PVA are LA volume, LA diameter, and non-paroxysmal AF. In the present study, long-term freedom from any AT/AF was achieved in 84% after 1 year including 1.7 ± 0.6 PVA (Figure 2B). This acts in concert with the reported success rates from other studies.10,1419 In this context, it has been previously reported that RMN-guided circumferential PV ablation with an open-irrigated catheter provides comparable acute and long-term success rates when compared with manual catheter navigation.14,15,17,18

LA imaging

Initially, imaging of the LA to guide PVA was performed using fluoroscopy alone, and this method is still the only visualization of LA and the PV ostia within PVA in several centres. In the last decade, several new imaging techniques came to clinical practice.20 In addition, gated MDCT and/or cardiac magnet resonance (CMR) images can be acquired before PVA to provide detailed information of the LA and to include these data into electroanatomic mapping systems, which offer 3D reconstructions of the LA to guide the physician within the procedure (Figure 1).6 Prior studies have compared MDCT favourably with other imaging methods including fluoroscopy, TEE, intracardiac echocardiography (ICE), and CMR.21,22 MDCT showed comparable diagnostic value to assess the LA anatomy, PV anatomy, as well as additional PVs, without user-depended problems with ICE, TEE, and fluoroscopy. It is a major limitation of MDCT that it is associated with a non-negligible radiation dose23 with scanners not using advanced dose reduction techniques. The technical development has led to newer MDCT scanners that help to reduce the total effective radiation dose to below 1–3 mSv.24,25 All patients included in this study received their ECG-gated MDCT for the purpose of imaging the LA and to measure LA volume prior to the PVA as a clinical standard procedure. Consequently, no additional radiation exposure was incurred. As an alternative imaging modality, CMR does not involve the use of ionizing radiation. However, while CMR provides images with excellent temporal resolution, it is time-consuming and even more expensive than some other imaging modalities. In patients with arrhythmia, image quality from gated MRI might be insufficient for reliable volumetry of the LA. Patients with severe heart failure may have problems to stay through the long-lasting imaging. Patients suffering from severe claustrophobia will not be able to tolerate the CMR scan. In addition, patients with implanted devices usually cannot undergo CMR.21,22

Figure 1

Left atrial (LA) volume evaluation by 64-MDCT. (AC) Assessment of LA volume using oblique axial planes. The black lines depict manual tracings of the LA, excluding the PV (A and C) and the LAA (C). (B) LA and the mitral annulus (*) as the border between LA and LV. (D) Respective three-dimensional rendering of the LA and PV (LAA, left atrial appendage; LSPV, left superior PV; LIPV, left inferior PV; RSPV, right superior PV; RIPV, right inferior PV), and the calculated LA volume (E).

Predictors of AT/AF recurrence

Several studies reported about well-known predictors of AF/AT recurrence and PVA success rates. A significant dilatation of the LA is thought to be a risk factor of large extent of LA remodelling, which might be related with a limitation of PVA success.8,9 However, various studies identified risk factors for early recurrence of AT/AF after PVA (e.g. type of AF, LA-dimensions).8,9,11,20 In our patient group, the LA diameter assessed by TTE was one of the main indicators for AT/AF recurrence after successful PVA (Tables 1 and 2, Figure 2C). Despite this, it was not a significant predictor for AT/AF recurrence in multivariate analysis. Figure 3 demonstrates a relatively low correlation of LA diameter (TTE) with LA volume measured by gated MDCT. Figure 3 indicates that discrepancy between the two imaging methods increases with LA enlargement. The plot shows a moderately good regression in patients with LA diameter <46 mm and LA volume <106 mL shown by the regression line in the graph. In contrast, the correlation weakens for LA diameters >45 mm and LA volumes >106 mL as illustrated in the graph by a more scattered distribution towards the right. However, we suggest that the LA enlargement refers to the level of structural remodelling of the LA. In this context, it has to be noted that LA volumetry, especially in patients with LA enlargement >106 mL, seems to be a better criterion for the success of the procedure in these patients than the anterior–posterior diameter. Table 2 shows that the type of AF was another significant predictor of AT/AF recurrence, which has been already confirmed by previous studies.12,26

Figure 2

(A) Freedom from any AT/AF recurrence during follow-up after the initial PVA procedure. The first 3 months are regarded as the blanking period. (B) Freedom from any AT/AF recurrence during follow-up including additional PVA procedures in the case of AT/AF recurrence after the index procedure. (C) Freedom from any AT/AF recurrence during follow-up for patients dichotomized by the median LA diameter of 45 mm (P = 0.063) and (D) LA volume of 106 mL (P = 0.024).

Figure 3

Scatter plot and linear and exponential correlation between LA diameter (TTE) and LA volume (MDCT). A relatively low linear correlation was demonstrated between both imaging modalities (P < 0.001, r = 0.51), an exponential fit yielded an association of P < 0.001 and r = 0.52. As shown in the plot, the discrepancy between the two imaging methods increases with LA enlargement.

Prognostic value of LA volume measurement

There are different smaller studies with various measurement protocols reporting the prediction of AT/AF recurrence at LA volumes range from >89 to >145 mL.12,2731 Our data of 368 procedures, using a manual measurement of the ECG-gated MDCT slices (Figure 1), indicate that the volume of ≥106 mL was a good point for discriminating those patients with AT/AF recurrence from those without. In this context, the AUC for the LA volume was 0.73 (P < 0.001). Concordantly, Figure 2D shows a significantly lower AT/AF-free probability in patients with an LA volume above the median LA volume than in those with an LA volume below (P = 0.024, Figure 2).

These results do absolutely agree with a prior study of only 99 patients by Abecasis et al.28 They report that the probability of relapse was significantly higher in patients with LA volumes >100 mL when assessed by MDCT. However, they used a semi-automatic software with atrium endocardial contour, automatic detection, and operator correction to evaluate the LA volume. To our knowledge, there is one other additional study operation manually to visualize the LA volume.12 This study indicates that patients with AF recurrence after the first ablation procedure had LA volumes >135 mL. To our opinion, a cut-off point close to the median LA volume seems to be ideal. In this context, it has to be considered that the sum of pre-procedural markers and patients comorbidities influence the outcome of PVA and not a single marker like LA enlargement alone.

Accuracy of LA volumetry by MDCT

It is still a matter of debate whether there is clinically relevant utility between different modalities to measure the LA prior PVA. Especially, whether there is benefit of obtaining MDCT instead of TTE to better predict the outcome of the PVA. To the best of our knowledge, this is the first prospective study that aimed to evaluate LA dimensions by different imaging modalities in a large patient group with 368 ablation procedures. LA diameter by TTE had a rare correlation with MDCT volume classification, especially in patients with severely enlarged LA. Furthermore, the AUC for the LA diameter measured by TTE was only 0.6 with an asymptotic significance in the ROC analysis of 0.09, respectively. The one-dimensional TTE measurement underestimates the dimensions visualized by MDCT and this leads to a lower predictive power of LA diameter. This could probably be explained by several reasons. Inaccuracies of a one-dimensional measurement of LA size have been well described12,27,28 and LA area from TTE was not a significant marker in multivariate analysis in a recent study.31 The real LA size can be significantly underestimated with TTE, and LA enlargement does often not occur in a single dimension. In addition, regarding the possibility of two-dimensional-derived LA volumes provided by TTE, it has to be considered that this method calculates a suggested LA volume and cannot directly measure the LA volume as MDCT does. Furthermore, imaging of the posterior wall of the LA was often difficult to achieve and likely still undersized the measurement of the LA volume. A major advantage of MDCT is the possibility to visualize the entire LA with reliable and reproducible imaging (Figure 1).

Clinical implication

To our opinion, pre-procedural ECG-gated MDCT provides a feasible tool for an observer-independent, objective, three-dimensional visualization of LA structural remodelling. We conclude that the investigation of LA volume might help the electrophysiologist to improve patient selection and avoid potential unnecessary procedures with the risk of unexpected complications. In addition, LA volume helps the operator to predict patients in whom successful PVA can be reached using circumferential isolation (LA volume <106 mL and small LA diameter) and those in whom a higher risk of AT/AF recurrence maybe expected and additional targets like linear lesions on the LA roof or the mitral-isthmus line become additional options.


LA volumetry by MDCT, in patients presenting for PVA with the help of MDCT reconstructions for electroanatomical mapping, is a good predictor of AT/AF recurrence. It can easily be included in the clinical pre-procedural practice.

Conflict of interest: none declared.


  • Both authors contributed equally to this work


View Abstract