Medial arterial calcification in the feet of diabetic patients and matched non-diabetic control subjects

Abstract

The prevalence and distribution of medial arterial calcification was assessed in the feet of four subject groups; 54 neuropathic diabetic patients with previous foot ulceration (U), median age 60.5 (50.5-67 interquartile range) years, duration of diabetes 19.5 (9.9-29.9) years; 40 neuropathic diabetic patients without a foot ulcer history (N), age 68 (62-73) years, duration of diabetes 14.0 (8.0-28.0) years; 43 non-neuropathic diabetic patients (NN), age 60.5 (52-68.5) years, duration of diabetes 14.0 (8.0-28.0) years and 50 non-diabetic control subjects, age 62.5 (53.7-70) years. A single radiologist graded medial arterial calcification as absent, mild or severe, at the ankle, hind-foot, mid-foot, metatarsals and toes on standardised plain lateral and antero-posterior foot radiographs taken by a single radiographer. Diabetes history, vibration perception threshold, ankle systolic pressure and serum creatinine were also assessed. Medial arterial calcification was significantly greater (total score 18 [3-31]) in neuropathic diabetic patients with previous ulceration (U vs N p<0.01, U vs NN p<0.001). Non-neuropathic diabetic patients did not have significantly higher arterial calcification scores than age-matched non-diabetic control subjects. Medial arterial calcification correlated with vibration perception threshold (r=0.35), duration of diabetes (r=0.32) and serum creatinine (r=0.41), (all p<0.01). Logistic regression models showed vibration perception and duration of diabetes to predict the probability of any calcification. Serum creatinine level was added to predict severe calcification. Ordered categorical modelling confirmed that medial arterial calcification was significantly heavier at the ankle than the toes for all groups, odds ratio 4.35 (2.94-6.43, 95% confidence intervals), (p<0.01). Ankle systolic pressure and ankle-brachial pressure index were significantly associated with degree of arterial calcification, r=0.40 and r=0.35, respectively, (both p<0.01) in diabetic patients. However, arterial calcification was present in more than one-third of patients with an ankle-brachial pressure index of less than 1.0. In conclusion, although ankle pressures correlate with the degree of arterial calcification, medial arterial calcification may be present in patients with low ankle systolic pressures, which may be falsely elevated even at 'normal' values. This finding may provide a rationale for the use of toe rather than ankle pressure measurements in diabetic patients, particularly those with peripheral neuropathy, and this hypothesis should be directly tested in future studies. © 1993 Springer-Verlag.