Abstract: SA-PO869

Mortality Prediction of Abdominal Aortic and Pelvic Calcification on Plain X-Ray in CKD, Hemodialysis, and Kidney Transplant Patients

Session Information

  • Vascular Calcification
    November 04, 2017 | Location: Hall H, Morial Convention Center
    Abstract Time: 10:00 AM - 10:00 AM

Category: Mineral Disease

  • 1205 Vascular Calcification

Author

  • Disthabanchong, Sinee, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Phayathai, Thailand
Background

Vascular calcification (VC) is highly prevalent in CKD and predicts poor outcomes. Both cardiovascular and CKD related risk factors participate in the development of VC. The gold standard for evaluation of VC is computed tomography but plain x-rays offer a less costly and less radiation exposure alternative. Lateral abdominal and pelvic x-rays have been utilized to evaluate the calcification of abdominal aorta, iliac and femoral arteries which was found to be highly correlated with abdominal aortic calcification (AAC) and CAC obtained by CT. The data regarding the predictability of AAC and pelvic arterial calcification (PAC) on patient outcomes are limited.

Methods

Four hundred and nineteen CKD stages 2-5 (CKD 2-5), maintenance HD (HD) and long-term KT (KT) patients were included. Only KT recipients who had been transplanted for at least 1 year were enrolled in order to allow the time for stabilization of mineral metabolism. AAC score as described by Kauppila et al and PAC score as described by Adragao et al were applied to determine the severity of AAC and PAC on lateral abdominal and pelvic x-rays respectively. The median follow-up time was 62.7 months for CKD 2-5, 52 months for HD and 62.5 months for KT.

Results

AAC and PAC scores correlated well with the correlation coefficients (r) >0.4 in all 3 populations (p<0.001). Patients with AAC score >6 or PAC score >1 were older, had higher prevalence of DM, serum PO4 and PTH but lower serum albumin and eGFR. Increased KT duration was associated with a more severe degree of AAC, whereas prolonged dialysis vintage was associated with a more severe degree of PAC. Kaplan-meier survival curves revealed AAC score >6 as a significant predictor for all-cause mortality in CKD 2-5 but not in HD or KT, whereas PAC score >1 was a significant predictor of mortality in all 3 populations. After an adjustment for age, the predictability of AAC in CKD 2-5 was lost, whereas PAC >1 remained an independent predictor of mortality in all 3 populations. Further adjustments for age, sex, BMI, serum albumin, calcium and PO4 revealed the predictability of PAC for mortality in CKD 2-5 patients and KT recipients but not in HD patients.

Conclusion

PAC was a better predictor of mortality than AAC in all 3 populations of CKD and should be considered in the evaluation of VC in CKD and KT patients.