Autonomic function testing in diabetic subjects using sequential measurements.

The repeatability of commonly used electrocardiographic (ECG) related autonomic function tests was investigated in 25 diabetic subjects, mean age 44 (range 18-67) years and mean duration of diabetes 10 (< 1-35) years. Tests were based on deep breathing, the Valsalva manoeuvre, relaxed normal breathing, and standing up from a lying position. All tests were repeated twice at each session, with two sessions separated by a mean of 5 (range 3-8) months. ECG and respiratory data were recorded for subsequent analysis. Maximum (max) and minimum (min) RR intervals and instantaneous heart rate (HR) were determined. The mean and repeatability data (within-subject standard deviation) for 21 different test results were calculated, and included deep breathing sitting (max-min)RR 226 +/- 39 ms and (max-min)HR 19.3 +/- 3.2 beats min-1, Valsalva manoeuvre 1.74 +/- 0.18, and lying-to-standing RR ratio 1.19 +/- 0.07. The within-subject repeatability to between-subject variability ratios consistently demonstrated that it is better to perform the deep breathing test sitting (with the smaller mean ratio of 0.37) rather than supine (0.46), with the use of three respiratory cycles rather than a single cycle (0.33 compared with 0.50 for (max-min)RR). A significant (p = 0.037) relationship was found between variability in heart rate changes and variability in air volume breathed during the deep breathing test.