BACKGROUND
Diabetic autonomic neuropathy is a relatively common diabetic complication, associated with high long-term mortality. Ewing's test is known as the 'gold standard' for evaluating and diagnosing this disease, yet is not widely used due to the inconvenient procedures of the test. 24-hour Holter EKG monitoring, and the analytical product, heart rate variability, is being introduced as a relatively simple and reliable procedure for the evaluation of diabetic autonomic neuropathy. We explored whether such heart rate variability products derived from Holter monitoring, correlated with the presence, absence, or severity of diabetes mellitus, and whether it correlated well with conventional autonomic tests. METHODS: We compared 59 type 2 diabetic patients with 71 normal subjects. All underwent 24-hr Holter EKG monitoring and basic autonomic evaluations, such as the head-up tilting, hand grip, and deep breathing-heart rate variability tests. Those who had diabetes also underwent evaluation for basic blood chemistry, and complication studies, for things such as: 24-hour urine albumin excretion, fundoscopy and nerve conduction. RESULTS: Variables for heart rate variability were expressed as SDDN, rMSSD, LF, HF, and LF/HF, where SDDN is the Standard Deviation of all RR intervals, rMSSD the square root of the mean of the sum of the squares of differences between adjacent RR intervals, LF the power in the Low Frequency range and HF the power in the High Frequency range, with LF/HF being the ratio between LF and HF. Heart rate variability was significantly lower in terms of rMSSD, LF, HF, but not in terms of the LF/HF ratio, for the diabetic patients compared to the normal subjects. These three variables also correlated with the conventional autonomic tests of systolic blood pressure changes during standing up (negatively), and heart rate variability during deep breathing (positively). SDDN, rMSSD, LF, and HF also correlated negatively with the duration of diabetes. SDDN, LF and HF were significantly lower among patients who had complications such as: retinopathy, nephropathy or peripheral neuropathy, than in those who did not. CONCLUSION: Heart rate variability was lower in type 2 diabetic patients than the control subjects, which correlated well with the duration of diabetes mellitus, diabetic chronic complications and the conventional autonomic nervous function tests, so could be an useful adjunct or even a replacement, for conventional autonomic nervous system testing procedures. More research is needed in this field.