Hide and Seek: Does the TBI Allow for Earlier Recognition of PAD in Diabetic Patients?

Abstract

Introduction: It is assumed that in diabetic patients, calcification of the ankle arteries may cause unreliable measurement of the anklebrachial index (ABI). Clinical guidelines recommend the toebrachial index (TBI) as an alternative in diabetic patients with ‘falsely elevated’ ankle pressures, arbitrarily defined as an ABI >1.4. Considering that arterial calcification is also common among diabetics with an ABI <1.4, and may thus result in a ‘falsely normal’ ABI and subsequent under-diagnosis of PAD. We investigated whether diabetics have a lower TBI at similar ABI as compared to non-diabetics, and if the TBI may enable earlier detection of PAD in diabetics. Methods: We randomly selected 326 diabetic and non-diabetic patients (512 legs) with suspected PAD from our vascular lab registry. Mean difference between ABI and TBI was compared for diabetics and non-diabetics. In addition, a Bland-Altman plot was constructed with 95% limits of agreement established from nondiabetics. Separate analyses were performed including only patients with Fontaine stages 2 or 3, or an ABI within the normal reference range (0.91e1.4). Results: Diabetic and non-diabetic patients were similar with regard to age and sex distribution. Median ABI did not differ between both groups (Table 1). Median TBI was higher in diabetics, but overall the difference between ABI and TBI was similar. Remarkably, in patients with Fontaine 2 or 3, mean difference between ABI and TBI was larger for non-diabetics (mean difference 0.11, 95% CI 0.20 to 0.03; p 1⁄4 0.008). Among patients with a normal ABI, both the TBI and the difference between ABI and TBI were similar for diabetics and non-diabetics (Table 1). The difference between ABI and TBI for diabetics overlapped the reference range established from non-diabetics, independent of the magnitude of the measurements (Figure 1). Conclusion: We found no indication that the TBI may enable earlier detection of PAD in diabetics. The TBI and ABI are strongly correlated, and this relation is not influenced by the presence of diabetes. In patients with Fontaine 2 or 3, TBIs were actually lower in non-diabetics at similar ABIs. As such, initial assessment of the TBI in diabetics, compared to non-diabetics, generally does not yield additional information if the ABI is not obviously elevated.