Bedside diagnosis of cardiac autonomic damage by computerized analysis of heart rate-respiration relationship

Early autonomic dysfunction in patients with diabetes mellitus assessed by spectral analysis of heart rate and blood pressure variability

Patients with diabetes mellitus (DM) often have alterations of the autonomic nervous system (ANS), even early in their disease course. Previous research has not evaluated whether these changes may have consequences on adaptation mechanisms in DM, e.g. to mental stress. We therefore evaluated whether patients with DM who already had early alterations of the ANS reacted with an abnormal regulatory pattern to mental stress. We used the spectral analysis technique, known to be valuable and reliable in the investigation of disturbances of the ANS. We investigated 34 patients with DM without clinical evidence of ANS dysfunction (e.g. orthostatic hypotension) and 44 normal control subjects (NC group). No patients on medication known to alter ANS responses were accepted. The investigation consisted of a resting state evaluation and a mental stress task (BonnDet). In basal values, only the 21 patients with type 2 DM were different in respect to body mass index and systolic blood pressure. In the study parameters we found significantly lower values in resting and mental stress spectral power of mid-frequency band (known to represent predominantly sympathetic influences) and of high-frequency and respiration bands (known to represent parasympathetic influences) in patients with DM (types 1 and 2) compared with NC group (5.3 +/- 1.2 ms2 vs. 6.1 +/- 1.3 ms2, and 5.5 +/- 1.6 ms2 vs. 6.2 +/- 1.5 ms2, and 4.6 +/- 1.7 ms2 vs. 6.2 +/- 1.5 ms2, for resting values respectively; 4.7 +/- 1.4 ms2 vs. 5.9 +/- 1.2 ms2, and 4.6 +/- 1.9 ms2 vs. 5.6 +/- 1.7 ms2, and 3.7 +/- 2.1 ms2 vs. 5.6 +/- 1.7 ms2, for stress values respectively; M/F ratio 6/26 vs. 30/14). These differences remained significant even when controlled for age, sex, and body weight. However, patients with DM type 2 (and significantly higher body weight) showed only significant values in mental stress modulus values. There were no specific group effects in the patients with DM in adaptation mechanisms to mental stress compared with the NC group. These findings demonstrate that power spectral examinations at rest are sufficiently reliable to diagnose early alterations in ANS in patients with DM. The spectral analysis technique is sensitive and reliable in investigation of ANS in patients with DM without clinically symptomatic autonomic dysfunction.

Decreased heart rate variability in patients with chronic obstructive pulmonary disease

To evaluate possible autonomic nervous system (ANS) dysfunction in patients with chronic obstructive pulmonary disease (COPD) in the absence of any hypoxic neuronal damage, we studied 31 patients with COPD patients aged 31 to 68 years (55 +/- 10) and 32 age-matched healthy subjects (control). Respiratory function in the patients was as follows: FEV1 = 52 +/- 8 percent; PaO2 = 71 +/- 14 mm Hg; and PaCO2 = 40 +/- 10 mm Hg. The ANS was assessed by heart rate variability (HRV) in the time domain (SD of mean RR interval) and frequency domain (autoregressive spectral analysis recognizing low [LF] and high [HF] frequency components, vagal and sympathetic related, respectively). Patients and controls were evaluated at rest and during vagal (controlled breathing [CB]) and sympathetic (passive head-up tilt) maneuvers. Patients with COPD showed a depressed global HRV (rest SD = 34 +/- 20 ms vs 45 +/- 15 ms, p < 0.05; tilt SD = 28 +/- 14 ms vs 38 +/- 13, p < 0.01) with a predominant respiratory drive (rest HF = 44 +/- 28 vs 28 +/- 18, p < 0.05; tilt HF 42 +/- 28 vs 16 +/- 12, p < 0.01) as compared with normal subjects. In the control group, vagal and sympathetic responses were in opposite directions following a stimulus, whereas there was no significant HRV response in the COPD group. We conclude that patients with COPD have abnormalities of ANS function, with in particular a depressed HRV response to sympathetic and vagal stimuli.

Phasic changes in human nasal and skin blood flow: relationship with autonomic tone

To assess the autonomic control of the microcirculation in human skin and nasal mucosa, the spontaneous fluctuations of a laser Doppler signal recorded on forearm skin and nasal mucosa were compared with heart rate fluctuations and the respiratory signal in 11 healthy subjects by means of spectral analysis. Possible direct mechanical-thermal effects of airflow on the nasal mucosa were evaluated by comparing recordings obtained when the subjects were breathing through either the nose or the mouth. Comparison of the spectral analysis of the signal obtained for skin and nasal mucosa blood flow allowed us to define the characteristics of the autonomic tone of the microcirculation in these two areas. The skin is mainly influenced by sympathetic activity, and the nasal blood flow is regulated by both sympathetic and parasympathetic activity.

Impaired circadian modulation of sympathovagal activity in diabetes. A possible explanation for altered temporal onset of cardiovascular disease

BACKGROUND

Diabetic subjects have a high incidence of cardiovascular accidents, with an altered circadian distribution. Abnormalities in the circadian rhythm of autonomic tone may be responsible for this altered temporal onset of cardiovascular disease.

METHODS AND RESULTS

To assess circadian changes of sympathovagal balance in diabetes, we performed 24-hour power spectral analysis of RR interval fluctuations in 54 diabetic subjects (age, 44 +/- 2 years) with either normal autonomic function or mild to severe autonomic neuropathy and in 54 age-matched control subjects. The power in the low-frequency (LF, 0.03-0.15 Hz) and high-frequency (HF, 0.18-0.40 Hz) bands was considered an index of relative sympathetic and vagal activity, respectively. Diabetic subjects with autonomic abnormalities showed a reduction in LF compared with control subjects (5.95 +/- 0.12 In-msec2 versus 6.73 +/- 0.11, p < 0.001) and an even greater reduction in LF, particularly during the night and the first hours after awakening (5.11 +/- 0.18 In-msec2 versus 6.52 +/- 0.14, p < 0.001). Day-night rhythm in sympathovagal balance was reduced or absent in diabetic subjects compared with control subjects.

CONCLUSIONS

Diabetic subjects with or without signs of autonomic neuropathy have a decreased vagal activity (and hence a relatively higher sympathetic activity) during night hours and at the same time of the day, during which a higher frequency of cardiovascular accidents has been reported. These observations may provide insight into the increased cardiac risk of diabetic patients, particularly if autonomic neuropathy is present.

Autonomic nervous system and microcirculation in diabetes

In order to evaluate the occurrence of sympathetic impairment of skin microvascular control in diabetes, we evaluated the spectral analysis of forearm skin laser-Doppler fluctuations in nine insulin-dependent diabetic subjects and in 21 controls of similar age. Low-frequency oscillations (around 0.1 Hz) were significantly lower in diabetics than in controls (2.333 +/- 0.340 (mean +/- SEM) units vs. 3.486 +/- 0.093 units, P less than 0.001), whereas no significant differences were found between the two groups regarding high-frequency respiration-related oscillations. These results suggest that the loss of rhythmicity in diabetic subjects is selectively related to low-frequency oscillations, mostly under sympathetic control, and is likely to be dependent on autonomic abnormality.

Relationship between retinal ischemia and cardiovascular autonomic damage in diabetes

In order to investigate the relationship between ischemic diabetic retinopathy and cardiac autonomic dysfunction we studied 40 diabetic patients (mean age +/- SEM: 48 +/- 2 years). Retinopathy was evaluated by fluorescein angiography and patients were divided into three groups: a) with ischemic retinopathy (15 patients), b) with edematous retinopathy (13 patients) and c) without retinopathy (12 patients). Cardiac autonomic function was assessed by 4 tests based on cardiovascular reflexes (heart rate response to Valsalva maneuver and to deep breathing; systolic blood pressure fall after standing and diastolic blood pressure rise during handgrip) and by the cross-correlation test, a computerized technique for the analysis of respiratory sinus arrhythmia based on spectral analysis of electrocardiographic and respiratory signals. Abnormalities of cardiac autonomic tests were present in 26/40 patients (65%); considering the retinal angiographic classification, autonomic test abnormalities were found in 13/15 (86.6%) patients with ischemic retinopathy, in 7/13 (53.9%) patients with edematous retinopathy and in 6/12 (50%) patients without retinopathy. The prevalence of cardiac autonomic dysfunction was significantly higher (p less than 0.005) in the ischemic retinopathy group compared to the other groups. These results show a strong association between ischemic retinopathy and cardiac autonomic dysfunction.

Relationship between phasic changes in human skin blood flow and autonomic tone

Heart rate beat-to-beat oscillations synchronous with respiration and blood pressure waves, have been found to be a marker of sympathovagal interaction in man and animals. Oscillations of heart rate, respiration, and cutaneous blood flow were simultaneously recorded to assess the relationship between autonomic nervous control and cutaneous circulation in a group of 21 healthy subjects and in a group of 6 healthy patients after brachial plexus anesthesia and consequent sympathetic blockade. In the first group, changes in posture were employed to modify autonomic tone. Relative changes in cutaneous blood flow were recorded by laser-Doppler flowmetry. Spectral analysis techniques (cross-correlation) were used to quantify the relationship between oscillations common to the recorded signals. A standing maneuver induced a significant decrease of the cross-correlation between respiratory and heart rate fluctuations (from 4.93 +/- 0.16 to 4.44 +/- 0.16 a.u.; P less than 0.001), and a significant increase of the cross-correlation between heart rate and skin blood flow fluctuations (from 0.64 +/- 0.31 to 1.33 +/- 0.21 a.u.; P less than 0.001), but did not modify the cross-correlation between respiratory and skin blood flow fluctuations (from 2.87 +/- 0.15 to 3.04 +/- 0.14 a.u.; NS). After the standing maneuver the maximum correlation between heart rate and skin blood flow was always due to oscillations in the range of 0.1 Hz (or 10-sec period), similar to the oscillations described in large arteries. Sympathetic blockade reduced significantly the cross-correlation between heart rate and skin blood flow (P less than 0.001). These results suggest that the cross-correlation between skin blood flow and heart rate at 10-sec period fluctuations can be used as an index of the influence of the autonomic tone on skin blood circulation.

Relationship between fluctuations in heart rate and asymptomatic nocturnal ischaemia

In order to quantify autonomic changes related to asymptomatic nocturnal myocardial ischaemia, we analyzed heart rate fluctuations recorded during Holter monitoring in 9 subjects with coronary heart disease (21 episodes) and in 11 age-matched controls. R-R interval spectral analysis was computed in sequences of 256 heart beats, taken during the ischaemic episode, 4, 8 and 60 minutes before, and 4 and 60 minutes after. Mean heart rate, R-R interval variability (assessed by R-R interval standard deviation), low and high (respiration-linked) frequency components of R-R interval spectrum were evaluated. Mean heart rate and R-R interval variability increased only during ischaemia (from 62.9 to 73.3 beats/minute, P less than 0.02, and from 39 to 88 msec, P less than 0.01, respectively). While high-frequency components of heart rate variability remained unchanged, low-frequency peak increased during ischaemia (from 9.4 to 43.3 sec2 X 10(-3)/Hz, P less than 0.01) and also 8 minutes (P less than 0.05) and 4 minutes before (P less than 0.05). Despite a moderate increase of heart rate occurring only during ischaemia, the early rearrangement of heart rate fluctuations suggests the occurrence of changes of autonomic tone before the electrocardiographic onset of ischaemia. Due to its limited amount, this phenomenon appears to be a consequence, most likely unspecific, of factors responsible for the genesis of myocardial ischaemia.