Heart rate variability in diabetic patients during orthostatic load--a spectral analytic approach

Autonomic Nervous Function and Blood Volume Monitoring during Hemodialysis

We studied the autonomic reflex response to hypovolemia during HD by means of spectral analysis of heart rate variability (SAHRV) in 10 hypotension prone (group A) and 10 hemodynamically stable patients (group B). UF rate normalized per total body water and blood volume fall were similar in the two groups. The sympatho-vagal balance index, calculated as the ratio between the integrals of the spectrum in the low (0.02-0.15 Hz) and in the high frequency range (0.2-0.35 Hz) rose in group B progressively from the beginning of the treatments, reaching a top at the 90th minute and remained subsequently high until the end. On the contrary this index did not show remarkable increases in group A. The differences between the two groups were statistically significant at 0 (4.6∓2.9 vs 1.5∓1.3), 60th (8.3+7.8 vs 2.2+2.6), 90th (17.9+13.4 vs 3.7+2.8), 150th (8.8+3.7 vs 2.7+2.8) and 210th minute (8.6+5.0 vs 2.9+2.5). In conclusion SAHRV shows an impairment of autonomic reflex response to hypovolemia in hypotension-prone patients.

Time- and frequency-domain estimation of early diabetic cardiovascular autonomic neuropathy

The risk related to cardiovascular autonomic neuropathy dysautonomia should lead to a specific assessment of this complication of diabetes. The aim of this study was to estimate the accuracy of a battery of blood pressure (BP) and heart rate (HR) variability indexes obtained in different subgroups of diabetic subjects classified according to the conventional laboratory autonomic function tests (Ewing scores). Blood pressure was measured continuously at the finger level with a Finapres monitor while subjects were in the supine position and again while they were standing. Pulse intervals were derived from BP recordings and were taken as surrogates for R-R intervals. Subjects with borderline or definite cardiovascular autonomic neuropathy showed a similar degree of alterations of both HR and BP variability (spectral measures) and in the relationship between BP and HR (cross-spectral and sequence analysis). Subjects with no evidence of cardiovascular autonomic neuropathy on the basis of the conventional tests showed an altered relationship between BP and HR. This baroreceptor-HR reflex dysfunction could represent an early stage of cardiovascular autonomic neuropathy undetected by the conventional tests. The areas under the receiver operating characteristic plots indicated that the high-frequency peak of pulse interval was highly discriminant in the supine and standing positions. The cross-spectral analysis showed the best discrimination for the gain in the high-frequency range. For the sequence analysis, the slope was the best discriminant factor for any degree of cardiovascular autonomic neuropathy. In conclusion, these estimates of baroreceptor-HR function may provide a powerful tool for assessing cardiovascular autonomic neuropathy at any stage, including the early stage, which is not detected by the conventional tests.

Heart rate variability and baroreceptor function in chronic diabetic rats

In conscious chronic (12 to 18 weeks) streptozotocin diabetic rats, we examined the changes in basal heart rate, with particular attention to heart rate variability assessed by evaluating the standard deviation (bpm) of the lengths of adjacent pulse pressure. We also investigated in anesthetized rats the ability of the aortic baroreceptors to acutely (30 minutes) reset to hypertensive levels. For this purpose, pressure-nerve activity curves for the baroreceptors were obtained, and gain (slope of the curve) and mean arterial pressure at 50% of maximal baroreceptor activity were calculated. The shift of the pressure-nerve activity curve was used as an index of resetting. Conscious diabetic rats (n=6) exhibited lower mean arterial pressure (93+/-6 versus 109+/-4 mm Hg), heart rate (272+/-25 versus 359+/-11 bpm), and heart rate variability (18+/-7 versus 36+/-6 bpm) than control rats (n=7). Under anesthesia, diabetic rats (n=7) and control rats (n=8) exhibited similar mean arterial pressure (113+/-6 versus 109+/-7 mm Hg in control rats ), mean arterial pressure at 50% of maximal baroreceptor activity (117+/-5 versus 107+/-6 bpm), gain (1.66+/-0.08 versus 1.81+/-0.05%/mm Hg), and extent of resetting (44+/-12 versus 49+/-9%) to hypertensive levels. The present study demonstrated that conscious chronic diabetic rats presented lower heart rate variability than control rats. On the other hand, chronic diabetes was not associated with alterations in baroreceptor function or its ability to rapidly reset to hypertensive levels.

Power spectral analysis of heart-rate variability reflects the level of cardiac autonomic activity in rabbits

Power spectral analysis of heart rate (HR) variability was tested in conscious rabbits to assess the reliability of this method for assessing cardiac autonomic function in normal rabbits under resting conditions. Evaluation of power spectrum was performed in 5 rabbits under normal resting conditions and after sympathetic, parasympathetic and combined sympathetic plus parasympathetic blockade. Rabbits were randomly assigned to undergo sympathetic (propranolol) or parasympathetic (methscopolamine) blockade at the initial step followed by combined blockade. The power spectrum of heart-rate variability in rabbits was presented as one broad spectral component at frequencies mainly between 0 and 0.5 Hz. This component was considerably modulated by both sympathetic and parasympathetic influences with substantial overlap of sympathetic- and parasympathetic-related components of the spectrogram. Nevertheless, it was clearly shown that power of heart-rate variability at frequencies from 0.4373 Hz to 0.5625 Hz was determined only by parasympathetic influences, and sympathetic modulation of HR was presented mainly at frequencies from 0.0625 Hz to 0.1875 Hz. Spectral subcomponent analysis of the power spectrum of HR variability may be useful to follow changes in cardiac autonomic function in rabbits.

Predictors of the pressor response to the clinic visit in essential hypertensives with and without diabetes mellitus

Predictive models for the pressor response to the outpatient clinic visit (PRC) in essential hypertensives with and without diabetes are proposed. The hypotheses are derived from previous studies about the univariate correlates of this response. PRC was measured with ambulatory monitors. Twenty-four hour blood pressures and average PRCs were similar in the two groups. Diabetics had faster 24-h heart rates, decreased heart rate variability, a broader range of PRCs and more depressor responders. PRC of nondiabetics correlated with duration of hypertension and was dependent on race; the predictive model had R2 of 0.19. In contrast, PRC of diabetics exhibited correlations with age, weight, BP and blood glucose and the model had R2 of 0.71. The data suggest that: diabetics had autonomic dysfunction, that their PRC can be modelled with predictors that are accepted correlates of autonomic neuropathy, and that these predictors attenuated PRC or its buffering. If these results were confirmed by prospective application of the model to a larger group of patients, 'true' blood pressures could be estimated by subtraction of predicted PRC from office blood pressures in diabetic, but not in nondiabetic, hypertensive patients.

Prevalence of cardiovascular autonomic dysfunction assessed by spectral analysis, vector analysis, and standard tests of heart rate variation and blood pressure responses at various stages of diabetic neuropathy

To establish a test battery for the detection and characterization of cardiovascular autonomic neuropathy (CADN) and to evaluate its prevalence, a number of autonomic function tests based on spectral analysis, vector analysis, and standard tests of heart rate variation and blood pressure responses were performed in 261 diabetic patients aged 11-76 years with various stages of peripheral neuropathy. The percentages of abnormal results in the individual tests based on heart rate variation were 6-31% in 115 patients without peripheral neuropathy, 16-45% in 61 patients with subclinical neuropathy, 22-59% in 73 patients with symptomatic peripheral neuropathy, and 67-100% in 12 patients with the latter in conjunction with autonomic symptoms (p < 0.05). The most frequently abnormal indices, each representing a different physiological basis, were the coefficient of variation, low-frequency and mid-frequency power spectrum at rest, mean circular resultant, postural change in systolic blood pressure, and, in particular, the max/min 30:15 ratio and Valsalva ratio. CADN, defined as the presence of > or = 3 abnormalities among these seven parameters was detected in none of 120 control subjects, 13.0% of the patients without peripheral neuropathy, 34.4% of those with subclinical neuropathy, 49.3% of those with symptomatic peripheral neuropathy, and in 100% of the subjects with the latter and concomitant autonomic symptoms (p < 0.05). The overall prevalence of CADN in 103 patients completing all parameters was 46.6%. The corresponding rate of CADN defined as > or = 2 abnormalities among the five tests included in an optimized version of the battery proposed by Ewing and Clarke was 38.8%.(ABSTRACT TRUNCATED AT 250 WORDS)

Age-related changes of heart rate power spectra in a diabetic man during orthostasis

We used spectral analysis of heart rate variability during orthostatic load in diabetic patients in order to provide quantitative information about first, the alterations of vagal and beta-adrenergic nervous system influence on the heart and secondly, the relation of spectral components of heart rate variability to age. The respiration-related power, and indirect estimate of vagal cardiac control, was significantly reduced in diabetics as compared with controls in both the supine (diabetics: 0.39 (0.18-1.15) l/min2 [median and 25-75% percentiles], controls: 1.52 (0.54-3.84) l/min2, 2P less than 0.01) and the standing (diabetics: 0.27 (0.12-0.84) l/min2, controls: 2.22 (0.67-5.14) l/min2, 2P less than 0.0001) body posture. The difference between supine and standing values of mid-frequency power (delta MF power) that represents beta-adrenergically mediated heart rate variability was significantly smaller in diabetics (delta MF: 0.79 (0.18-3.84) l/min2) than in controls (delta MF: 8.07 (4.35-17.99) l/min2) (2P less than 0.0001). Power of almost all frequency components declined exponentially with age in both positions and groups studied. The decline in power of heart rate variability was more pronounced in controls than in diabetics for either posture. An age-associated significant decrease in delta MF was found in controls (Y = 33.16-0.60X, P less than 0.05, r = -0.49) as compared to diabetics (Y = 8.70-0.16X, P greater than 0.05, r = -0.24). Our results suggest a decreased vagal and sympathetic nervous system input into the heart in diabetics. The premature loosened coupling between heart rate spectral power and age suggests that diabetes mellitus accelerates the aging process. Computerized spectral analysis of heart rate variability seems to be a very sensitive tool to evaluate the influence of the autonomic nervous system on the heart.