Relationship between Early and Late Recurrence
In the Non-CF group, 32 of 54 patients (59.3%) with ER developed LR. The proportion of patients with ER was significantly greater among patients with LR than among those without LR (76.2% vs. 40.0%, p < 0.001). There was no difference in the rate of a single ER between patients with and without LR (23.3% vs. 18.5%, p = 0.373), whereas the proportion of patients with ER occurring in multiple months were significantly greater in patients with LR than in those without LR (51.2% vs. 22.2% p = 0.001). Freedom from LR in patients with the last ER in the 1st or 2nd month was significantly greater than in those with the last ER in the 3rd month (p < 0.001; Figure 4a).
Similarly, in the CF group, patients with LR had a higher rate of ER than those without LR (84.6% vs. 21.5%, p < 0.001). However, in the CF group, the proportions of patients with ER in both single month and multiple months were significantly greater in patients with LR than in those without LR (single ER: 47.7% vs. 20.2%, p < 0.001; multiple ER: 36.9% vs. 1.3%, p < 0.001). The rate of freedom from LR in patients with the last ER occurring in the 1st and 2nd months was 69.8% and 40.0%, respectively, whereas all patients with the last ER occurring in the 3rd month had LR (Figure 4b).
When comparing patients with the last ER occurring in the 3rd month in the CF and Non-CF groups, the CF group was significantly less likely to be free from LR at 2 years post-procedure than the Non-CF group (Figure 4c). The significance of ER as a risk factor of LR is summarized in Table 3. Occurrence of the last ER in the 1st or 2nd month was not a significant risk factor for LR in the Non-CF group, whereas any ER at any time-point was a significant risk factor for LR in the CF group. Of note, occurrence of the last ER in the 3rd month was an absolute risk factor for LR in the CF group.