Abstract: SA-PO862
Peripheral Vascular Calcification Score Is Better Than Abdominal Aortic Calcification Score of Plain Radiographs in Predicting Cardiovascular Mortality in Hemodialysis Patients
Session Information
- Dialysis: Cardiovascular, BP, Volume
October 27, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 701 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Mo, Min, Shunde Hospital Southern Medical University, Foshan, China
- Pan, Jianyi, Shunde Hospital Southern Medical University, Foshan, China
- Zhang, Wei, Shunde Hospital Southern Medical University, Foshan, China
- Zheng, Qingkun, Shunde Hospital Southern Medical University, Foshan, China
- Chen, Jinzhong, Shunde Hospital Southern Medical University, Foshan, China
- Dou, Xianrui, Shunde Hospital Southern Medical University, Foshan, China
Background
Cardiovascular disease (CVD) is the major cause of death for dialysis patients. Vascular calcification (VC) serves as a key pathological factor resulting in the CVD. KDIGO advocates abdominal aortic calcification score (AACS) to evaluate VC. Meanwhile, some studies showed that the peripheral vascular calcification score (PVCS) could predict cardiovascular mortality. However, no studies compare two VC score systems in clinic. Thus, this study compared two VC score systems in predicting mortality of HD patients.
Methods
In this retrospective study with 243 HD patients, AACS and PVCS were measured from plain radiograph of lateral abdominal radiograph and pelvis and both hands radiographs, respectively.
Results
The prevalence of VC was 68.5% (167 patients), most patients showed AAC (63.7%). During the follow-up period of 24 (13,44) months, 65(26.6%) patients died. Among these died patients, 35 (53.8%) patients died of cardiovascular disease. The patients died showed higher AACS (5.0 (1.25,14.0) vs. 2.0 (0,6.75), P=0.001), PVCS (2.0(0,5) vs. 0(0,2.0),P=0.000), and overall VC score ( 5 (0,16) vs. 1 (0,6), P=0.001) compared to alive patients. When ROC curve was used to predict all-cause mortality, for PVCS, the area under ROC curve (AUROC) was 0.695 with 57.4% of specificity and 71.8% of sensitivity (P=0.000). As for AACS, the AUROC was 0.666 with 62.9% of specificity and 61.7% of sensitivity (P=0.001). The AUROC for PVCS was 0.668 with 68.0% of specificity and 62.5% of sensitivity (P=0.007), while the AUROC for AACS was 0.574 with 38.1% of specificity and 75.0% of sensitivity (P=0.239) when used to predict cardiovascular mortality. By the COX multivariate regression analysis (adjusted for age, gender, diabetes, period of dialysis, albumin, hypertension, cholesterol, phosphorus, calcium, intact-parathyroid and Kt/V), the PVCS was independently associated with all-cause mortality and cardiovascular mortality (HR, 1.287;95%CI, 1.140-1.453, and HR, 1.272;95%CI,1.079-1.499), respectively. However, AACS was only associated with all-cause mortality (HR, 1.069; 95%CI, 1.008-1.133) but not cardiovascular mortality.
Conclusion
VC score is independently associated with mortality. The PVCS is a better parameter in predicting cardiovascular mortality as compared with AACS.