Upper limb vascular calcification score as a predictor of mortality in diabetic hemodialysis patients

This study evaluated the correlation between an upper limb vascular calcification (Vc) score (VcS) and late all-cause mortality in diabetic hemodialysis patients with distal upper limb arteries medial wall sclerosis (Mönckeberg disease).

We retrospectively reviewed Vc in bilateral upper limb plain radiographs and in duplex ultrasound images performed before radial-cephalic fistula (RCF) creation in diabetic hemodialysis patients. Only medial linear calcifications outlining the vessel wall were considered positive on X-ray images, whereas for ultrasound reviews, only continuous highly echogenic plaques producing bright white echos with shadowing were considered to be medial calcification. A VcS was then applied in each patient. Every half of each of the three main arterial conduits (brachial, radial, and ulnar arteries) in each arm was counted as 1 if it contained ≥ 6 cm of linear calcification, whereas absence of calcification or minimum calcification (length <6 cm) was counted as 0. Long-term all-cause mortality was compared between patients with a low or moderate VcS <8 (group I), patients with a high VcS ≥ 8 (group II), and patients with VcS = 0 (control group). Kaplan-Meier statistics were used for comparisons among the groups.

Nineteen patients had a VcS <8, 21 had VcS ≥ 8, and 43 patients had VcS = 0. The study patients had a mean age of 68 ± 10 years (range, 42-83 years; P = .23). Before early conversion to a RCF, dialysis therapy in 59 (71.1%) had already been initiated through central venous catheters (CVCs). The mean follow-up for groups I, II, and controls was 41.4 ± 41.2 months (range, 4-144 months), 34.15 ± 31.3 months (range, 1-108 months), and 66.7 ± 32.5 months (range, 12-126 months), respectively (P = .0009). Forty-seven patients died during the follow-up period (12 in group II and 24 in the controls; P = .88). Survival rates at 12, 24, 36, and 48 months were 78.3%, 65.7%, 54.8%, and 48.1% for group I; 75.2%, 58.8%, 49.3%, and 42% for group II; and 97.7%, 93.1%, 76.8%, and 71.8% for the control group, respectively (P = .013 for all groups; P = .044 for group II vs controls). Patients with (subgroups) or without CVCs at baseline had similar late mortality rates. Patients with CVCs/Vc had lower survival rates than those with CVCs/no Vc at 1 year (73.3% vs 96.5%) and at 3 years (47.7% vs 75.8%; P = .038). CVCs were related to increased risk of death only in subgroup II patients compared with the subcontrol group patients (75.4% vs 37.9% at 5 years, respectively; P = .034).

Diabetic hemodialysis patients exposed to high levels of upper extremity arterial medial VcSs upon receiving RCFs have an increased long-term mortality risk compared with diabetic hemodialysis patients with no Vc and receiving the same access. Patients with CVCs/Vc had the lowest survival rates.