Discussion
This was a prospective observational study of the characteristics of the ACZ in patients with AF using the databases from two centers. The main findings were:
(1) the ACZ was observed in 95% patients, and all ACZ was linearly distributed,
(2) the ACZ was most frequently observed in the anterior wall region (77%),
(3) a longer ACZ was significantly associated with a larger LA size and a prevalence of non-PAF,
(4) a larger LA size was associated with a higher overlapping rate of LVA and ACZ, and
(5) a larger LA size was not associated with a higher overlapping rate of CoAs and ACZ.
With the development of the 3-D mapping system, there has been significant focus on the conduction pattern of the LA. Recently, Yamaguchi et al reported that a difference in bipolar voltage mapping between ring catheter and HD Grid.14 They stated that the HD Grid can create a voltage map that is independent of propagation direction and can assess more precise LA conduction pattern. To date, several studies have reported on the conduction abnormalities of the LA. However, the conduction abnormalities varied between patients, and there were various methods of mapping and different definitions of the ACZ in each study.
A limited number of studies have reported on conduction abnormalities at the LA in humans. Markides et al studied the LA activation during SR in 19 AF patients with noncontact mapping system. They found that line of conduction block running vertically in the LA posterior wall was observed in all patients. Moreover, they reported that the anterior line extending from the anterior mitral valve annuls toward the left atrial appendage was observed in 9 patients (47%).15 Mouws et al investigated the conduction abnormalities during sinus rhythm only at the LA posterior wall using epicardial mapping in 268 patients who underwent cardiac surgery.6 They showed that conduction abnormalities occurred in 90% of patients. Moreover, in their study, AF episode was associated with a higher incidence of conduction abnormality, larger number of lines, and longer lines. Roberts-Thomson et al reported the conduction properties of the posterior LA during SR using epicardial mapping in 34 patients who underwent cardiac surgery.16 In their study, 5 patients (15%) had an ACZ running vertically down the posterior LA. The results of our study showed that the ACZ was observed in 95% patients, and all ACZ was linearly distributed, similar to the results of previous studies. On the other hand, the prevalence of the ACZ was higher in our study than those in previous studies although our study did not include patients with structural heart disease. This discrepancy in findings might be due to different patient inclusion criteria, methods of mapping (epicardial or endocardial mapping), mapping catheter, and definition of the ACZ. However, very few studies have focused on the ACZ using the HD Grid. Further investigation on the ACZ is warranted.
In our study, the ACZ was associated with a non-PAF, BMI, hypertension, and LA diameter/BSA. These factors are well known predictors which increase the prevalence of AF. It has also been established that non-PAF increases predisposition to LA remodeling in a so-called “AF begets AF” manner. A recent study showed that patients with AF had a higher incidence of conduction abnormality, larger number of lines, and longer lines than those without AF.6 These findings support our results indicating that existence of ACZ is associated with non-PAF, BMI, hypertension, and LA diameter/BSA. Moreover, we found that a longer ACZ was significantly associated with a larger LA size. The association between LA size and ACZ yielded mixed results in previous studies. Roberts-Thomson et al reported that the line of conduction delay was most marked in conditions associated with greater LA enlargement.16 In contrast to our results, the other study showed that LA dilation was not predictor of long line of conduction abnormalities.6Furthermore, their patients had a larger LA size compared with our patients because they included structural heart disease such as valvular heart disease. This discrepancy in findings may be explained by the different baseline LA size.
It is well-known that the ACZ plays an important role in atrial tachyarrhythmias, however, the causes of the ACZ are still unclear. We considered the following mechanisms of the ACZ. First, the ACZ might be the result of an abrupt change in myocardial fiber direction. A previous study showed that all hearts have subendocardial fibers, and the longitudinal fibers were arranged in broad bands. Moreover, they found that the most obvious broad band of longitudinal fibers was formed by the “septopulmonary bundle”.17 Markides et al also reported that abnormal conduction was associated with a change in these myocardial fiber direction.15 In fact, the location of the ACZ shown in Figure 1 matches the septopulmonary bundle at the anterior wall. On the other hand, the incidence of the ACZ in the posterior wall in our study was lower than in the previous study. This discrepancy in the incidence of the ACZ may be explained by the different activation flow. In our study, the activation in the posterior wall was activated vertically from the roof and bottom because mapping was performed during high right atrial pacing (Supplementary video ). Therefore, the incidence of the ACZ in the posterior wall was low in our study, and the pacing from different site might have led to the different distribution of the ACZ. Second, the compression from the external organs might induce the ACZ. Mayyas et al found that endothelin-1, which promote myocyte hypertrophy and interstitial fibrosis, is an important factor in AF development and persistence.18 They also mentioned that greater wall stress may produce higher levels of endothelin-1 because this factor is stretch mediated. Based on this report, we hypothesized that externally-mediated wall stress leads to regional differences in the ACZ. In our study, the overlapping rate of LVA (mild and moderate) and ACZ was higher than that of CoAs and ACZ. Therefore, the ACZ is associated with an LVA rather than a CoAs. These findings indirectly suggest that the ACZ might be a precursor to LVA.