Background and aims: 

Diabetic foot problems are among the most serious consequences of diabetes mellitus. Lower limb amputations in diabetic patients represent almost 50% of all non-traumatic lower extremity amputations. Upper extremities are affected much less frequently than the lower. While the role of peripheral obliterative macroangiopathy in the pathogenesis of diabetic gangrene is well established, the role of microangiopathy is less equivocal. The rhythmic variations of skin microvascular blood flow reflect the influence of heartbeat, respiration, intrinsic myogenic activity, neurogenic factors and endothelial activity. The aim of our study was to test the hypothesis that basal skin blood flow (BSBF) and its dynamic components differ 1) among diabetic patients and non-diabetic control subjects without large vessel disease and with it, and 2) among the upper and lower extremities.

Patients and methods: 

69 diabetic patients (D) and 78 control subjects (C) entered the study. Peripheral arterial obliterative disease (PAOD) was defined according to ankle/brachial index (ABI). All subjects were classified into three groups: 1 (PAOD +): ABI< 0.9 (21 D, 16 C); 2 (PAOD -): ABI 0.91 – 1.3 (43 D, 36 C); and 3 (mediocalcinosis): ABI > 1.3 (5 D, 26 C). BSBF at 4 recording sites with predominantly nutritive capillary circulation (right and left caput ulnae, right and left medial malleolus) was measured by laser Doppler flowmetry (LDF). The oscillatory components of the LDF signal were analyzed by wavelet transform.

Results: 

In absolute terms, mean flow at all recording sites was highest in group 2, but the differences among the groups were not statistically significant. The values in the upper extremities were significantly higher than in the lower, except for groups D3 and C1. In addition, significant difference was found between the two arms in D2 and between the two legs in C1 (Table 1). In D, the spectral component of microvascular flow associated with endothelial activity was in significant positive correlation with systolic pressures on brachial and dorsal pedal artery (p=0.001 and 0.010, respectively).

Conclusion: 

Our results indicate that mean BSBF and its oscillatory components do not change with diabetic PAOD; however there is a strong correlation between systolic pressure and the oscillatory components of BSBF related to endothelial activity manifested in the frequency interval 0.0095 – 0.02 Hz. The differences between the upper and lower extremities provide a possible explanation for the higher prevalence of ulceration and gangrene on the lower extremities in comparison to the upper.

Table 1: Mean flow at 4 recording sites in diabetic patients and control subjects.