Background  

Because dipyridamole is used to assess heart rate (HR) variability, we investigated whether a low HR response during dipyridamole single photon emission computed tomography (SPECT) in patients with diabetes indicates the presence of cardiac autonomic neuropathy (CAN).

Methods and Results  

Subjects were 61 non-insulin-dependent diabetes patients without perfusion defects, myocardial infarction, or arrhythmia who underwent thallium 201 SPECT imaging. The control group comprised 28 subjects without diabetes. HR was measured during infusion of dipyridamole at a rate of 0.14 mg/kg/min, and peak-baseline ratios of 1.20 or less were defined as low. CAN severity was classified by standard autonomic function tests as severe (n = 22), mild (n = 19), or none (n = 20). HR ratios were significantly attenuated in patients with diabetes compared with those in control subjects (1.22 ± 0.12 vs 1.32 ± 0.12, P <.001). Among the patients with diabetes, HR ratios decreased as CAN severity increased from none (1.32 ± 0.10) to mild (1.23 ± 0.12, P <.05) to severe (1.13 ± 0.08, P <.005). There was good correlation between HR ratio and R-R interval ratio to deep breathing and to Valsalva, and patients with low HR ratios showed an attenuated response to both tests (all P <.001). The sensitivity and specificity of HR ratios in the detection of CAN were 77% and 74% for severe CAN and 63% and 90% for mild-to-severe CAN, respectively.

Conclusions  

In patients with diabetes who have normal dipyridamole SPECT results, an attenuated HR response observed during stress indicates a high likelihood of CAN. Further work that assesses these results in diabetes patients with coronary artery disease is warranted.