Region-specific analysis of acute PVR
In a total of 620 segments other than SAE and 124 SAE, acute PVR was
observed in 43 (7%) and 21 (17%) segments (p = 0.001), respectively.
The detailed distribution of acute PVR sites is depicted in the Figure
2. In SAE, there was only a significant difference in ILDmax between the
segments with acute PVR and those without; whereas in the segments other
than SAE, there was a significant difference in CFmin, FTImin, AImin,
and Imp-min (Table 2). In the ROC curve analysis, ILDmax and Imp-min had
the highest area under the curve (AUC) to predict the absence of acute
PVR in the SAE and the segments other than SAE, respectively (0.68 and
0.69) among various 3-D mapping related indices.
ILDmax of 4.8 mm discriminated durable segments with a specificity of
90% in the SAEs and also in the entire segments. Conversely, an Imp-min
of 6.5 ohm could distinguish segments with gaps and no gaps with a
specificity of 90% in segments other than SAE. In entire-segment
analysis, AI min of 383 au was required to discern durable lesion with a
specificity of 90%, while AI of 405 au for the segments other than SAE
and AI of 334 au for the SAE were related to the absence of acute PVR
with a specificity of 90%.