Ctors for all-cause and cardiovascular mortality. In contrast, older age was independently associated only with all-cause mortality (Table 2).Table 2. Multivariate Cox’s proportional hazard models of baseline aortic arch calcification (AoAC) all-cause and cardiovascular mortality.All- cause mortality HR Age (years) Male gender 95 CICardiovascular mortality HR 95 CI 0.988?.069 0.254?.206 0.389?.PPNS NS NS1.048 1.022?.074 ,0.001 1.028 1.136 0.660?.954 NS 0.554 0.772 3.807 0.522 1.453 1.002 0.Diabetes mellitus 1.071 0.679?.690 NS Cardiovascular disease 2.000 1.143?.500 0.1.441?0.054 0.007 0.226?.209 0.688?.071 0.972?.032 0.389?.285 1.044?.996 1.577?.132 NS NS NS NS 0.034 0.History of smoking0.928 0.520?.657 NS Lipid-lowering therapy Ca6P (mg2/dL2) Albumin (g/dL) 1.027 0.629?.676 NS 0.989 0.970?.007 NS 0.763 0.520?.118 NSLog hs-CRP (mg/L)1.725 1.257?.367 ,0.001 1.769 Baseline AoAC 2.181 1.336?.561 0.002 3.Progression of AoAC: Subgroup Analysis According to the Presence of Baseline AoACFollow-up chest 
X-rays at 12 months after PD start were available in 363 patients. Among them, 140 patients (38.5 ) had AoAC at baseline and 223 patients (61.5 ) did not. The progression of AoAC was significantly more observed in patients with AoAC at baseline (P,0.001). Among 140 patients with AoAC at baseline, 90 patients (64.2 ) experienced AoAC progression, whereas AoAC progressed in only 12 (5.3 ) out of 223 patients without baseline AoAC. Two hundred eleven patients with AoACS of zero at baseline remained free of AoAC during the 12-month follow-up. Pearson’s correlation analysis revealed that changes in AoACS were significantly associated with baseline AoACS (r = 0.389, P,0.001), age (r = 0.301, P,0.001), and time-averaged hs-CRP (r = 0.167, P = 0.001) and calcium concentrations (r = 0.124, P = 0.02). In multivariate binary logistic regression analysis,Ca, calcium; P, phosphate; hs-CRP, high sensitivity C-reative protein; HR, hazard ratio; CI, confidence interval; NS, not significant. doi:10.1371/ZK 36374 journal.pone.0048793.tbaseline AoACS (OR: 1.803, 95 CI: 1.383?.349, P,0.001), age (OR: 1.058, 95 CI: 1.016?.101, P = 0.006), and hs-CRP levels (OR: 1.904, 95 CI: 1.180?.070, P = 0.008) were found to be independent risk factors associated with AoAC progression. Since the baseline AoACS was significantly correlated with AoAC progression, subgroup analysis was performed to clarify the independent predictor for AoAC progression in patients with and without baseline AoAC. In patients with AoAC at baseline, there was a significant correlation between hs-CRP concentrations and the changes in AoACS (r = 0.248, P = 0.02), while changes in AoACS were significantly associated with age (r = 0.124, P = 0.04) and hs-CRP levels (r = 0.126, P = 0.036) in patents without baseline AoAC. However, the changes in Ca 6 P products andFigure 1. Kaplan-Meier analysis of (A) all-cause and (B) cardiovascular mortality in 415 patients. Patients with baseline aortic arch calcification (AoAC) showed significantly higher all-cause and cardiovascular mortality than those without (both log-rank test, P,0.001). doi:10.1371/journal.pone.0048793.gProgression of Aortic Arch Calcification in PDiPTH concentrations did not correlate with changes in AoACS in both NT 157 site subgroups. Similar findings were observed in binary logistic regression analysis. In patients with AoAC at baseline, univariate analysis reavealed that diabetes mellitus, previous cardiovascular disease, lipid-lowering therapy, hs-CRP lev.Ctors for all-cause and cardiovascular mortality. In contrast, older age was independently associated only with all-cause mortality (Table 2).Table 2. Multivariate Cox’s proportional hazard models of baseline aortic arch calcification (AoAC) all-cause and cardiovascular mortality.All- cause mortality HR Age (years) Male gender 95 CICardiovascular mortality HR 95 CI 0.988?.069 0.254?.206 0.389?.PPNS NS NS1.048 1.022?.074 ,0.001 1.028 1.136 0.660?.954 NS 0.554 0.772 3.807 0.522 1.453 1.002 0.Diabetes mellitus 1.071 0.679?.690 NS Cardiovascular disease 2.000 1.143?.500 0.1.441?0.054 0.007 0.226?.209 0.688?.071 0.972?.032 0.389?.285 1.044?.996 1.577?.132 NS NS NS NS 0.034 0.History of smoking0.928 0.520?.657 NS Lipid-lowering therapy Ca6P (mg2/dL2) Albumin (g/dL) 1.027 0.629?.676 NS 0.989 0.970?.007 NS 0.763 0.520?.118 NSLog hs-CRP (mg/L)1.725 1.257?.367 ,0.001 1.769 Baseline AoAC 2.181 1.336?.561 0.002 3.Progression of AoAC: Subgroup Analysis According to the Presence of Baseline AoACFollow-up chest X-rays at 12 months after PD start were available in 363 patients. Among them, 140 patients (38.5 ) had AoAC at baseline and 223 patients (61.5 ) did not. The progression of AoAC was significantly more observed in patients with AoAC at baseline (P,0.001). Among 140 patients with AoAC at baseline, 90 patients (64.2 ) experienced AoAC progression, whereas AoAC progressed in only 12 (5.3 ) out of 223 patients without baseline AoAC. Two hundred eleven patients with AoACS of zero at baseline remained free of AoAC during the 12-month follow-up. Pearson’s correlation analysis revealed that changes in AoACS were significantly associated with baseline AoACS (r = 0.389, P,0.001), age (r = 0.301, P,0.001), and time-averaged hs-CRP (r = 0.167, P = 0.001) and calcium concentrations (r = 0.124, P = 0.02). In multivariate binary logistic regression analysis,Ca, calcium; P, phosphate; hs-CRP, high sensitivity C-reative protein; HR, hazard ratio; CI, confidence interval; NS, not significant. doi:10.1371/journal.pone.0048793.tbaseline AoACS (OR: 1.803, 95 CI: 1.383?.349, P,0.001), age (OR: 1.058, 95 CI: 1.016?.101, P = 0.006), and hs-CRP levels (OR: 1.904, 95 CI: 1.180?.070, P = 0.008) were found to be independent risk factors associated with AoAC progression. Since the baseline AoACS was significantly correlated with AoAC progression, subgroup analysis was performed to clarify the independent predictor for AoAC progression in patients with and without baseline AoAC. In patients with AoAC at baseline, there was a significant correlation between hs-CRP concentrations and the changes in AoACS (r = 0.248, P = 0.02), while changes in AoACS were significantly associated with age (r = 0.124, P 
= 0.04) and hs-CRP levels (r = 0.126, P = 0.036) in patents without baseline AoAC. However, the changes in Ca 6 P products andFigure 1. Kaplan-Meier analysis of (A) all-cause and (B) cardiovascular mortality in 415 patients. Patients with baseline aortic arch calcification (AoAC) showed significantly higher all-cause and cardiovascular mortality than those without (both log-rank test, P,0.001). doi:10.1371/journal.pone.0048793.gProgression of Aortic Arch Calcification in PDiPTH concentrations did not correlate with changes in AoACS in both subgroups. Similar findings were observed in binary logistic regression analysis. In patients with AoAC at baseline, univariate analysis reavealed that diabetes mellitus, previous cardiovascular disease, lipid-lowering therapy, hs-CRP lev.