Disabling postural hypotension complicating diabetic autonomic neuropathy.

A 35-year-old Type 1 diabetic man with severe disabling postural hypotension was studied for physiological abnormalities, precipitating factors, and effect of current treatment. A 24-h blood pressure profile indicated a diurnal variation in systolic blood pressure with the lowest values recorded between 0100 and 0600 h, during which the patient often lost consciousness on standing (mean standing systolic pressure 78 mmHg at night vs 105 mmHg in the afternoon, p less than 0.001). Food induced a profound fall in systolic pressure, both while supine and while standing erect. The systolic pressure fall during euglycaemia was 49 mmHg vs 3 mmHg during hypoglycaemia. Plasma noradrenaline and adrenaline levels were low during euglycaemia, but increased during hypoglycaemia. Therapeutic manoeuvres aimed at increasing heart rate (by atrial tachypacing) and reducing the peripheral pooling of blood (vasoconstricting drugs and gravity suit), together with the somatostatin analogue octreotide, proved ineffective. These observations demonstrate the phenomenon of post-prandial exacerbation of postural hypotension in a Type 1 diabetic patient, and indicate that despite failure of conventional methods of treatment, hypoglycaemia increased plasma catecholamines and was effective in abolishing the blood pressure fall on standing.     

Classification of symptoms of hypoglycaemia in insulin-treated diabetic patients using factor analysis: relationship to hypoglycaemia unawareness.

The allocation of some symptoms of acute hypoglycaemia to autonomic and neuroglycopenic groups has proved problematical, with possible misinterpretation of studies which depend upon the use of diverse symptom questionnaires. Two hundred and ninety-five randomly selected insulin-treated diabetic patients were asked to report the symptoms which they usually experienced and believed to be caused by hypoglycaemia. Sweating, trembling, inability to concentrate, weakness, hunger and blurred vision were the most frequently reported symptoms. To classify symptoms of hypoglycaemia objectively, Factor Analysis was used to identify related symptoms which grouped together. This resulted in five groups or clusters of symptoms, four of which could be denominated as groups with a presumed common aetiology, and were labelled: 'neuroglycopenic', 'general malaise', 'autonomic', 'motor dysfunction', and 'sensory dysfunction'. The groups of symptoms derived by Factor Analysis were assessed in relation to partial or absent symptomatic awareness of hypoglycaemia based on historical evidence from the 295 insulin-treated diabetic patients. Neuroglycopenic symptoms were reported more commonly by the patients who had reported partial awareness of hypoglycaemia (number of symptoms 2.6 +/- 1.8 (mean +/- SD] than by the patients who had reported normal hypoglycaemia awareness (1.4 +/- 1.5 symptoms) (p less than 0.05). By contrast autonomic symptoms were reported less frequently by the patients who had reported absent hypoglycaemia awareness (1.3 +/- 1.4 symptoms) than by those with normal awareness (2.2 +/- 1.4 symptoms) (p less than 0.05), which was similar to the number of autonomic symptoms reported by the patients who had partial hypoglycaemia awareness (2.1 +/- 1.3 symptoms).     

自主神经功能与糖尿病患者左心室功能、心律失常的关系

目的 探讨糖尿病患者自主神经功能与部分心脏合并症之间的关系。方法 采用24小时动态心电图和血压同步监测102例糖尿病患者，左心室功能和结构参数用彩色多普勒超声心动图测定。结果 随SDNN（24小时内全部正常R－R间期的标准差）降低，糖尿病患者的舒张末期二尖瓣口血流频谱A峰与E峰峰值流速的比值（PVA／PVE）、室间隔厚度（IVSTd)、左心室后壁厚度（LVPWTd)、左心室重量指数（LVMI）均增加；二尖瓣前叶活动斜率（MEF）降低，IVSTd/LVPWTd则无变化；严重的心律失常如Lown‘s≥3级的室性心律失常及短阵房速、房颤的发生率也明显增加。上述变化与夜间血压的异常增高相伴随。结论 SDNN低的糖尿病患者心脏受累严重，并可能预示患者日后发生心源性猝死的危险性增加。     

Weak relationship between symptom perception and objective hypoglycaemia-induced changes of autonomic function in hypoglycaemia unawareness in diabetes

To assess the relationship between symptom perception and neurophysiological characteristics in hypoglycaemia unawareness, we investigated the awareness of symptoms, objective changes of autonomic function and counter-regulatory neuroendocrine responses to hypoglycaemia in intensively treated type I (insulin-dependent) diabetic patients with different degrees of hypoglycaemia unawareness. Hypoglycaemia (venous plasma glucose below 2.2 mmol/l) was induced with an intravenous insulin bolus in subjects with a history of repeated severe hypoglycaemia and hypoglycaemia unawareness (n = 10) and in a comparable group with good awareness of hypoglycaemia (n = 8). Autonomic symptoms, selected parameters of autonomic function and counter-regulatory hormones were assessed serially. Although hypoglycaemia was more pronounced in unaware patients (1.6 vs 2.0 mmol/l, P = 0.05), their induced adrenaline response was markedly impaired (delta adrenaline: 1.25±1.10 vs 2.55±1.46 nmol/l, P = 0.05). Astonishingly, differences between both patient groups in the course of autonomic function changes did not reach the level of significance (P = 0.35–0.92), although the unaware group reported markedly fewer autonomic symptoms, both neurogenic (P = 0.001) and neuroglycopenic (P = 0.04) than the aware group. This study indicates that in hypoglycaemia unawareness even extensive changes in autonomic function are not sufficient for the perception of hypoglycaemia and confirms that the central nervous system plays an important role in the awareness of hypoglycaemia. Received: 4 July 1997 / Accepted in revised form: 27 November 1997     

Further evidence for a high incidence of nocturnal hypoglycaemia in IDDM: no effect of dose for dose transfer between human and porcine insulins

We tested the hypothesis that transfer from porcine to human insulin causes a fall in nocturnal blood glucose and an increase in the frequency of hypoglycaemic episodes. Twenty IDDM patients (age 19–55, duration 3–36 years) used Velosulin and Insulatard twice daily for 12 weeks, double-blinded to species (human (H) or porcine (P)) in a randomized crossover study. Species was changed after 4 weeks’ run-in and 4 weeks later, with insulin doses unchanged on transfer. Ten patients underwent each sequence (H/P/H or P/H/P) and were admitted on the first and eighth night after transfer for hourly blood glucose measurement (22.00–07.00). Biochemical hypoglycaemia (<3.5 mmol l⁻¹) was observed on 39 of the 80 patient-nights studied (48.75 %). The number of episodes were similar during each night (H1 8, H8 10, P1 10, P8 11, p = 0.83). Total reported symptomatic episodes (H 51 vs P 73, p = 0.85), total HbA₁ (H 9.8 ± 0.3 %, P 10.0 ± 0.3 %, p = 0.32) and daily insulin doses (H 0.63 ± 0.04 units kg⁻¹ day⁻¹ vs P 0.63 ± 0.05 units kg⁻¹ day⁻¹, p = 0.54) were not different. Despite an apparent fall in blood glucose levels from night 1 to 8 on transfer to human (AUC 82.3 ± 7.8 vs 61.4 ± 5.3 mmol.h l⁻¹, p < 0.05) but not porcine insulin (AUC 70.7 ± 7.2 vs 70.1 ± 7.5 mmol.h l⁻¹, p = 0.74), there was no difference when all 4 nights were considered together (p = 0.30). We conclude that dose for dose transfer to human insulin does not increase numbers of episodes of nocturnal or reported hypoglycaemia. © 1997 by John Wiley & Sons, Ltd.     

Cell-mediated immunity and symptomatic diabetic autonomic neuropathy

T lymphocytes have been implicated in the nerve damage observed in allergic experimental neuritis and in idiopathic polyneuritis. Symptomatic autonomic neuropathy in long-standing Type 1 diabetes is a rare and unexplained complication, and some preliminary evidence has suggested a pathogenetic role for the immune system. We have measured levels of activated T lymphocytes in 18 Type 1 diabetic patients with symptomatic autonomic neuropathy and in 16 matched patients with uncomplicated Type 1 diabetes. Purified T lymphocytes from peripheral blood were stained with a fluorescein-labelled monoclonal antibody directed to the activation marker HLA-DR and counted under UV microscopy. Percent DR positive T lymphocytes were significantly raised in the patients with autonomic neuropathy when compared with long-standing uncomplicated diabetic patients (8.2 +/- 4.2 vs 4.9 +/- 3.3%, p less than 0.01). This finding lends support for a role of the immune system in the development of autonomic neuropathy.     

The prevalence of impaired glucose counter-regulation during an insulin-infusion test in insulin-treated diabetic patients prone to severe hypoglycaemia

Summary Among 603 patients over the age of 18, with insulin-treated diabetes mellitus, a questionnaire identified 98 patients who during a 12-month period had experienced severe hypoglycaemia, defined as an event which required the help of another person. Twenty of these patients had repeatedly suffered from such episodes, without any obvious reason, for a period of at least three years. The capacity to counter-regulate a standardized, insulin-infusion test (0.034 U·kg^–1·h^–1 given for 3 h unless severe neuroglucopenia developed) was evaluated in 14 of these patients. In 12, an impaired glucose counter-regulation was registered, defined as blood-glucose values below — 2 SD of healthy subjects. In all but one of these patients, a combined deficiency of glucagon and adrenaline was documented, and was believed to be the likely cause of their inclination towards hypoglycaemia. In patients with severe hypoglycaemia, but not in diabetic patients without severe hypoglycaemia or in healthy subjects, a significant relationship between insulin disappearance and glucose rise was found. It is concluded that in insulin-treated diabetic patients, the prevalence of recurrent attacks of severe hypoglycaemia amounts to about 4%. In such patients, a combined deficiency of adrenaline and glucagon responses to hypoglycaemia is the predominant finding and the disappearance rate of insulin becomes critical for recovery of blood glucose after hypoglycaemia.     

Symptomatic hypoglycaemia in 411 type 1 diabetic patients

The frequency of symptomatic hypoglycaemic episodes was studied in 411 randomly selected conventionally treated Type 1 diabetic out-patients. Between two consecutive visits to the out-patient clinic each patient filled in a questionnaire at home. The number of hypoglycaemic episodes was then recorded prospectively in a diary for 1 week. From the questionnaires, the (retrospective) frequencies of mild and severe symptomatic hypoglycaemia were 1.6 and 0.029 episodes patient-1 week-1. From the diaries, the (prospective) frequencies of mild and severe hypoglycaemic episodes were 1.8 and 0.027 patient-1 week-1. Symptomatic hypoglycaemia was more frequent on working days than during weekends (1.8:1) and more frequent in the morning than during the afternoon, evening, and night (4.5:2.2:1.4:1). The symptoms of hypoglycaemia were non-specific, heterogeneous, and weakened with increasing duration of diabetes. During their diabetic life, 36% of the patients had experienced hypoglycaemic coma. The frequency of hypoglycaemia was positively, but only weakly, correlated with insulin dose, number of injections, percentage unmodified insulin of the total dose, and HbA1c (mild hypoglycaemia only). The frequency was also negatively, but weakly, correlated with age and HbA1c (episodes with coma only), but not correlated with sex, duration of diabetes, or patients' ratings of worries about mild and severe hypoglycaemia.     

Is autonomic neuropathy a risk factor for severe hypoglycaemia? The EURODIAB IDDM Complications Study

Summary The hypothesis that diabetic patients with autonomic neuropathy are at increased risk of severe hypoglycaemia was examined in an epidemiological study of over 3000 IDDM patients in Europe (EURODIAB IDDM Complications Study). Autonomic function was assessed by two standard cardiovascular tests: change in heart rate and systolic blood pressure on standing. Severe hypoglycaemia was defined as an attack serious enough to require the help of another person. Compared to patients (68 %) reporting no attacks in the last year, those reporting one or more attacks were older (34.0 ± 10.7 vs 32.1 ± 9.9 years, mean ± SD, p < 0.0001), had had diabetes for a longer period (16.6 ± 9.5 vs 13.8 ± 9.1 years, p < 0.0001), had better glycaemic control (HbA_1c 6.4 ± 1.8 vs 6.9 ± 1.9 %, p < 0.0001) and were more likely (p = 0.002) to have abnormal responses to both autonomic tests (13.0 vs 7.7 %). A single abnormal autonomic response was not associated with an increased risk of severe hypoglycaemia. The odds ratio for severe hypoglycaemia in people with abnormal responses to both autonomic tests, compared to those with normal responses, was 1.7 (95 % confidence interval 1.3, 2.2) after controlling for age, duration of diabetes, glycaemic control and study centre. In conclusion, a combined autonomic deficit in heart rate and blood pressure responses to standing is associated with only a modest increase in the risk of severe spontaneous hypoglycaemia. Although the increase in risk is not large, severe hypoglycaemia was a frequently reported event in this study. IDDM patients with deficient autonomic responses who strive for tight glycaemic control may therefore be at particular risk of severe hypoglycaemia. [Diabetologia (1996) 39: 1372–1376] Received: 28 December 1995 and in final revised form: 11 June 1996     

Painful myocardial infarction in severe diabetic autonomic neuropathy

Summary Two cases are reported of painful myocardial infarction in diabetics with severe autonomic neuropathy confirmed by abnormal autonomic function tests. Painless myocardial infarction in diabetics has traditionally been attributed to damage of cardiac pain fibres by autonomic neuropathy but other factors such as microangiopathy in the myocardium may be responsible. It may simply be that diabetics come into hospital more often for other reasons and a silent myocardial infarction diagnosed incidentally.     

QT prolongation on the electrocardiogram in diabetic autonomic neuropathy

Patients with Type 1 diabetes and autonomic neuropathy have an increased risk of sudden death for which the mechanism remains obscure. Prolongation of the QT interval on the electrocardiogram may occur with sympathetic dysfunction and is also associated with ventricular arrhythmia and sudden death. We have therefore measured the QT interval in patients with Type 1 diabetes with normal, borderline, and definitely abnormal autonomic function tests and in non-diabetic control subjects. The maximum QT interval was measured on 12-lead electrocardiograms recorded at rest and then plotted against the RR interval. The QT interval was above the upper 95% limit for the non-diabetic control subjects in 5 diabetic patients with abnormal autonomic function tests (33%), but in no cases with normal or borderline tests. Multivariate analysis confirmed that autonomic score contributed significantly (p less than 0.025) to the variance in QT interval. The raw Valsalva ratio alone also contributed significantly to the variance in QT interval (p = 0.025). Heart rate variability, heart rate response to standing, age, sex, and the presence of symptoms of autonomic neuropathy did not contribute significantly.     

Epidemiological studies of diabetes mellitus in Denmark: 6. Use of hospital services by insulin-treated diabetic patients

Summary The use of hospital services among insulin-treated diabetic patients was studied in a group of 1499 patients, representing >98% of all prevalent cases as of 1 July 1973 in the Funen County, Denmark, who were followed during a 81/2 year period by a record linkage with the regional computerized hospital registration system. On the prevalence date, 26% of the patients (in the age group 0–9 years: 74%) attended a diabetic outpatient clinic. The overall average admission rates for males and females were 0.46 and 0.53 per diabetes-year, respectively, and the average estimated hospital bed-day occupancy rates per diabetes-year were 7.2 and 9.6, respectively. These figures are five times higher than expected from the general population. Diabetes was not recorded as a discharge diagnosis (primary and/or secondary) in 13% of the male and 15% of the female hospital bed-day occupancy rate.     

The time course and diurnal variation of postural hypotension in diabetic autonomic neuropathy

Nine diabetic patients (age range 23-67y) with autonomic neuropathy and symptomatic postural hypotension were studied. The time course of the postural drop in blood pressure (BP) was documented firstly at 0900h, prior to breakfast and insulin, and secondly in the afternoon. Seven patients were symptomatic in the morning but only one in the afternoon. The systolic nadir in the morning was lower (p less than 0.02) than in the afternoon. The mean systolic pressure in the first 3 min after standing was no different in the afternoon but it was lower in the 2nd (p less than 0.05), 3rd (p less than 0.02), and 4th (p less than 0.02) 3-min intervals in the afternoon. The time taken to reach the systolic nadir was variable, ranging from 1-12 min. It is concluded that the assessment of the postural fall in BP may be underestimated if measured immediately after standing or if it is measured in the afternoon.     

Insulin-induced hypoglycaemia and absorption of injected insulin in diabetic patients

The effect of insulin-induced hypoglycaemia (soluble insulin 1 mU kg-1 min-1 IV) on the absorption of 125I-labelled soluble insulin (10 U SC) from thigh was studied in 10 insulin-treated Type 1 diabetic patients on a test and a control day. Disappearance of 125I was followed by external gamma counting. Adipose tissue blood flow was measured concomitantly using the 133Xe-clearance technique. Arterial plasma levels of glucose, insulin, adrenaline and noradrenaline were determined intermittently. Hypoglycaemia occurred at a glucose level of 2.2 +/- 0.1 (+/- SE) mmol l-1 after 58 +/- 6 min. Peak levels of adrenaline (6.44 +/- 1.62 nmol l-1) and noradrenaline (2.29 +/- 0.39 nmol l-1) were found 10 min later. During the 30-min period after onset of hypoglycaemia, adipose tissue blood flow increased 132 +/- 45% (p less than 0.05) but the disappearance rate of 125I-insulin was unchanged. Thus, insulin absorption was unaltered in connection with hypoglycaemia in Type 1 diabetic patients, in contrast to the depression previously reported in healthy subjects, despite similar increases in adipose tissue blood flow.     

The problem of autonomic neuropathy in diabetic pregnancy

The hazards of pregnancy for both the mother and the fetus in diabetic women with severe retinopathy and nephropathy are well reported. We wish to highlight a poorly recognized problem in the obstetric management of the diabetic mother, that of pregnancy in a patient with autonomic neuropathy. Two such cases are reported where the presence of autonomic neuropathy severely jeopardized the health of the mother, with the loss of the fetus in one, due to occurrence of severe and intractable vomiting. The presence of moderate to severe symptomatic diabetic autonomic neuropathy, particularly with evidence of gastroparesis, may be a relative contraindication to pregnancy.     

Evaluation of periflux blood flow measurement in diabetic patients with autonomic neuropathy

The periflux blood flow (PBF) response pattern in the fingertips determined with a periflux laser Doppler flowmeter was analyzed in a standing test in which the subject was instructed to stand up from the sitting position and in Valsalva's maneuver in which respiration was held for 10 s at the maximum inspiration, and indicators such as reaction time (RT), orthostatic dysreaction rate (ODR) and PBF decreasing velocity (PDV) were established for determining autonomic abnormality. RT and ODR were significantly correlated with the change in blood pressure and pulse rate, respectively, in the standing test, and PDV was significantly correlated with beat-to-beat variation. Abnormalities were found in about half of the diabetics in the standing test and Valsalva's maneuver. These results suggested that the analysis of PBF changes in the standing and Valsalva's tests may be an easy and useful approach for clinical examination of the autonomic nervous system.     

Cholecystokinin and pancreatic polypeptide release in diabetic patients with and without autonomic neuropathy

The present study was undertaken to investigate postprandial responses of cholecystokinin (CCK) and pancreatic polypeptide (PP) and their interrelationship in patients with diabetes mellitus (DM) with and without autonomic neuropathy (AN). Twenty-two patients with DM (seven with AN and 15 without AN) and 14 age-matched healthy controls were studied. AN was diagnosed according to several tests of cardiovascular autonomic function. CCK and PP plasma levels were measured by specific radioimmunoassays before and at several time points after the oral administration of a test meal. Basal CCK plasma levels in DM patients were normal, whereas basal PP plasma levels were increased (139±18 vs 72±7 pg/ml;P<0.01). integrated=" postprandial=" cck=" response=" was=" increased=" in=" dm=" patients=" (208±27=" vs=" 110±14=" pmol/liter/2=">P<0.05), mainly=" due=" to=" the=" patients=" with=" an.=" postprandial=" pp=" response=" was=" increased=" in=" dm=" patients=" without=" an=" (37,273±5241=" vs=" 13,418±3299=" pg/ml/2=">P<0.001) but=" not=" in=" those=" with=" an=" (8887±3461=" pg/ml/2=" hr).=" moreover,=" pp=" response=" was=" closely=">P<0.002) correlated=" with=" the=" degree=" of=" an.=" a=" direct=" and=" linear=" correlation=" between=" postprandial=" cck=" and=" pp=" responses=" was=" found=" in=" healthy=" controls=">r=0.78;P<0.005) but=" not=" in=" dm=" patients.=" we=" conclude=" that=" the=" cck=" response=" to=" a=" meal=" is=" increased=" in=" diabetic=" patients=" with=" an,=" whereas=" the=" pp=" response=" is=" increased=" only=" with=" an=" intact=" autonomic=" nervous=" system.=" it=" is=" suggested=" that=" the=" correlation=" between=" postprandial=" hormonal=" responses=" in=" healthy=" subjects=" is=" due=" to=" the=" potency=" of=" cck=" as=" pp=" releasing=" agent=" and=" that=" this=" interaction=" does=" not=" work=" in=" diabetic=">     

QT interval prolongation in type 2 (non-insulin-dependent) diabetic patients with cardiac autonomic neuropathy

Summary QT interval alterations were measured in 41 non-insulin-dependent (type 2) diabetic patients and 14 age- and sex-matched control subjects. Cardiac autonomic neuropathy (CAN) was assessed by noninvasive tests (deep breathing, Valsalva maneuver and lying-to-standing) and diabetics were divided into three groups according to the results of these tests: diabetics with definitive (n=14), early (n=13) and without (n=14) CAN. The corrected values of QT intervals (QT_c) at rest were significantly longer in diabetics with definitive (447±5 ms; p<0.001), early (426±5 ms; p<0.05) and without (424±5 ms; p<0.05) CAN than in controls (407±5 ms). Moreover, QT_c intervals at rest were significantly (p<0.01) longer in diabetics with definitive CAN than in diabetics with early and without CAN. QT_c intervals at maximum tachycardia, induced by Valsalva maneuver, were considerably longer in diabetics with definitive CAN (451±6 ms) than in controls (407±6 ms; p<0.001) and in diabetics with early (434±6 ms; p<0.05) or without (422±6 ms; p<0.01) CAN. Furthermore, QT_c intervals at maximum tachycardia were significantly (p<0.01) longer in diabetics with early CAN than in controls. QT_c intervals at maximum bradycardia after Valsalva maneuver were significantly longer in diabetics with definitive (446±5 ms; p<0.001), early (434±5 ms; p<0.001) and without (424±5 ms; p<0.01) CAN than in controls (403±5 ms). Moreover, QT_c intervals at maximum bradycardia were considerably (p<0.01) longer in diabetics with definitive than without CAN. At least one abnormal (>440 ms) QT_c period was found in 19 out of 27 patients with early or definitive CAN, but 4 of 14 diabetics without any signs of CAN and none of the controls exibited abnormal QT_c period. It was concluded that QT_c interval prolongation due to imbalance of autonomic nervous tone could be observed in type 2 diabetic patients with CAN, suggesting a possible role in sudden cardiac death.     

Predictors of hypoglycaemia in insulin-treated type 2 diabetes patients in primary care: A retrospective database analysis

AimsTo investigate the frequency and predictors (diabetes care and treatment, comorbidity) of documented hypoglycaemia in primary care patients with insulin-treated type 2 diabetes.MethodsData from 32,545 patients (mean age: 70 (SD 11) years, 50.3% males) from 1072 practices were retrospectively analyzed (Disease Analyzer database Germany: 09/2011–08/2012). Logistic regression (≥1 documented hypoglyemia) was used to adjust for confounders (age, sex, practice characteristics, diabetes treatment regimen).ResultsThe prevalence of patients (12 months) with at least one reported hypoglycaemia was 2.2% (95% CI: 2.0–2.4%). The adjusted odds of having hypoglycemia were increased for renal failure (OR; 95% CI: 1.26; 1.16–1.37), autonomic neuropathy (1.34; 1.20–1.49), and adrenocortical insufficiency (3.08; 1.35–7.05). Patients with mental disorders including dementia (1.49; 1.31–1.69), depression (1.24; 1.13–1.35), anxiety (1.18; 1.01–1.37), and affective disorders (1.80; 1.36–2.38) also showed an increased odds of having hypoglycemia. Location of the practice in an urban area was associated with a lower odds ratio (0.74; 0.68–0.80).ConclusionsBoth individual patient characteristics (e.g. comorbidity) and regional factors (practice location) have a substantial impact on hypoglycaemia in primary care patients with insulin therapy.     

Cardiovascular autonomic neuropathy associated with carotid atherosclerosis in Type 2 diabetic patients.

AIMS: To clarify if cardiovascular autonomic neuropathy is associated with carotid artery atherosclerotic plaques in Type 2 diabetic patients. METHODS: Cardiovascular autonomic nerve function was related to carotid artery ultrasound in 61 Type 2 diabetic patients 5-6 years after diagnosis of diabetes. RESULTS: Cardiovascular autonomic neuropathy [abnormal age corrected expiration/inspiration (E/I) ratio or acceleration index (AI)] was found in 13/61 (21%) patients. Patients with cardiovascular autonomic neuropathy showed increased degree of stenosis in the common carotid artery (24.6 +/- 13.2% vs. 14.7 +/- 9.2%; P = 0.014) and a tendency towards a higher plaque score (4.0 +/- 1.7 vs. 3.2 +/- 1.6; P = 0.064). Controlled for age, AI correlated inversely with degree of stenosis (r = -0.39; P = 0.005), plaque score (r = -0.39; P = 0.005), and mean (r = -0.33; P = 0.018) and maximum (r = -0.39; P = 0.004) intima-media thickness in the common carotid artery. In contrast, E/I ratio correlated only slightly with mean intima-media thickness in the common carotid artery (r = -0.28; P = 0.049). CONCLUSIONS: Cardiovascular autonomic neuropathy was associated with carotid atherosclerosis in Type 2 diabetic patients. Abnormal E/I ratios reflect efferent structural damage to parasympathetic nerves whereas abnormal AI reflects afferent autonomic dysfunction possibly due to impaired baroreceptor sensitivity secondary to carotid atherosclerosis.