cardiac autonomic neuropathy, estimated ... - GNM Healthcare

Chronobiology International, 26(5): 942–957, (2009) Copyright # Informa Healthcare USA, Inc. ISSN 0742-0528 print/1525-6073 online DOI: 10.1080/07420520903044448

CARDIAC AUTONOMIC NEUROPATHY, ESTIMATED CARDIOVASCULAR RISK, AND CIRCADIAN BLOOD PRESSURE PATTERN IN DIABETES MELLITUS

Jose Cabezas-Cerrato, 1 Ramon Carmelo Hermida, 2 Jose Manuel Cabezas-Agrı´ cola,1 and Diana Elva Ayala2 1

Department of Medicine, University of Santiago de Compostela, University Hospital of Santiago and Endocrinology and Nutrition Service, Santiago de Compostela, Spain 2 Bioengineering & Chronobiology Labs., University of Vigo, Vigo, Spain

This study was designed to investigate potential factors involved in the disruption of the circadian blood pressure (BP) pattern in diabetes mellitus, as well as the relation between BP, cardiac autonomic neuropathy, and estimated cardiovascular risk. We studied 101 diabetic patients (58% with type 2 diabetes; 59% men), age 21–65 yrs, evaluated by 48 h BP monitoring. We performed three autonomic tests in a single session: deep breathing, Valsalva maneuver, and standing up from a seated position. Patients were classified according to the number of abnormal tests and their 10 yr risk of coronary heart disease or stroke. The prevalence of non-dipping 24 h patterning ranged from 47.6% in type 1 to 42.4% in type 2 diabetes. The awake/asleep ratio of systolic BP (SBP) was comparable between patients with or without abnormal autonomic tests. Pulse pressure (PP) was significantly higher in patients with 1 abnormal autonomic test (p , 0.001). Ambulatory SBP was significantly elevated in the group with higher risk of coronary heart disease (p , 0.001). Patients with higher stroke-risk had higher SBP but lower diastolic BP, and thus an elevated ambulatory PP by 9 mmHg, compared to those with lower risk (p , 0.001). Cardiac autonomic neuropathy is not the main causal-factor for the non-dipper BP pattern in diabetes mellitus. The most significant finding from this study is the high ambulatory PP found in patients with either cardiac autonomic dysfunction or high risk for coronary heart disease or stroke. After correcting for age, this elevated PP level emerged as the main cardiovascular risk factor in diabetes mellitus. (Author correspondence: [email protected]). Submitted January 15, 2009, Returned for revision February 20, 2009, Accepted April 01, 2009 This independent investigator-promoted research was supported in part by grants from Direcń General de Investigacio ń, Ministerio de Educacio ń y Ciencia (SAF2006-6254-FEDER); Consellerıá cio ńs Institucionais e Administracio ń Pu ń e ´ blica, Secretarıá Xeral de Investigacio de Presidencia, Relacio ń e OrdenaDesenvolvemento, Xunta de Galicia (PGIDIT03-PXIB-32201PR); Consellerıá de Educacio ń Universitaria, Direccio ń Xeral de Promocio ń Cientı´fica e Tecnolo ´xica do Sistema Universitario de cio ń e Industria, Direccio ń Xeral de InvestigaGalicia, Xunta de Galicia (RCH); Consellerıá de Innovacio ń, Desenvolvemento e Innovacio ń, Xunta de Galicia (INCITE07-PXI-322003ES; INCITE08-E1Rcio ń, University of Vigo. 322063ES); and Vicerrectorado de Investigacio Address correspondence to Jose Cabezas-Cerrato, Hospital Clıńico Universitario de Santiago, Travesia de la Choupana s/n, Santiago de Compostela, 15706, Spain. Tel.: 34 981 951245? Fax: 34 981/951546; E-mail: [email protected]

942

BP Pattern and Autonomic Neuropathy in DM

943

Keywords Diabetes mellitus, Circadian blood pressure variability, Cardiac autonomic dysfunction, Cardiovascular risk factors, Pulse pressure

INTRODUCTION Several attributes of the cardiovascular system, including blood pressure (BP) and heart rate (HR), exhibit predictable changes during the 24 h, generally in synchrony with the rest-activity cycle (Portaluppi & Smolensky, 1996). During the past two decades, specific features of the 24 h BP pattern have been assessed as potential sources of injury to target tissues and as triggers of cardiac and cerebrovascular events. Moreover, the prominent 24 h variation in the occurrence of a variety of acute cardiovascular events, such as myocardial infarction, angina pectoris, cardiac arrest, sudden cardiac death, and pulmonary embolism, have been shown to be closely related to the circadian BP pattern of hypertensive subjects (Portaluppi et al., 1999). A growing number of studies indicate the extent of the nocturnal BP decline is deterministic of cardiovascular injury and risk (Boggia et al., 2007; Brotman et al., 2008; Dolan et al., 2005; Ohkubo et al., 2002). These studies collectively show that the reduction of the normal 10 – 20% sleep-time BP decline characteristic of the non-dipper pattern is associated with elevated risk of target organ damage, particularly to the heart (left ventricular hypertrophy, congestive heart failure, and myocardial infarction), brain (stroke), and kidney (albuminuria and progression to end-stage renal failure). A blunted nocturnal BP decline is frequent among patients with diabetes mellitus (DM) (Chau et al., 1994; Fogari et al., 1993; Ikeda et al., 1993; Rutter et al., 2000; Wiegmann et al., 1990). The actual prevalence of a non-dipping or rising pattern (sleep BP mean . awake BP mean) in DM is, however, highly variable among different studies. Non-dipping in diabetic patients has been related to cardiac autonomic neuropathy (CAN) (Cardoso et al., 2008; Hornung et al., 1989; Monteagudo et al., 1996; Spallone et al., 1993, 2001, 2007). In addition, there is a broad consensus that CAN is a predictor of cardiovascular morbidity and mortality (Ewing et al., 1980; Maser et al., 2003). Whether this increased risk associated with CAN may be mediated by a disruption in the circadian BP pattern has not yet been clarified. Differences in classification criteria for both CAN and non-dipping have lead to uncertainties regarding the actual number and relative importance of the potential influential factors on circadian BP variability in DM. Accordingly, we investigated the potential relationship between the circadian BP pattern, CAN, and estimated cardiovascular risk in DM patients evaluated by 48 h ambulatory BP monitoring (ABPM).

944

J. Cabezas-Cerrato et al.

METHODS Study Design The study was conducted between October 2006 and July 2007 at the University Hospital of Santiago de Compostela, Spain. Diabetic patients of both types and sexes, 65 yrs of age, were recruited in a systematic way: one of every two diabetic patients attending each day the outpatient setting at the hospital. All patients had a diurnal activity/nocturnal resting routine; the average duration of nocturnal sleep was 8.8+0.1 h. Exclusion criteria included any type of arrhythmia or heart-block. Shiftworkers, heavy alcohol consumers ( 20 units/week), heavy smokers ( 15 cigarette/day), heavy exercisers, and persons of body mass index 45 kg/m2 were also excluded. Persons with cardiac, pulmonary, or renal insufficiency (creatininemia . 1.4 mg/dl), and those with any other known cause of cardiovascular autonomic dysfunction, were also excluded. Because there is some controversy (Stevens et al., 2006) regarding whether or not the decline of estimated glomerular filtration rate (eGFR) with age is normal, we decided to set 65 yrs as the upper limit as an inclusion criterion. This study was conducted in accordance with the World Medical Association’s fifth revision of the Declaration of Helsinki for research on humans and criteria set forth by the journal for ethical medical research (Portaluppi et al., 2008). The Institutional Review Board approved the study, and informed consents were obtained from all patients. Cardiorespiratory-reflexes (CR-R) tests are considered the closest to a “gold standard” method for CAN diagnosis. Tests were performed on each subject between 09:30 and 13:30 h in a quiet room with a controlled temperature (22 –248C). Deep breathing (DB), Valsalva maneuver (VM), and standing up (from a seated position) tests were performed by a computer-aided system (CASE-IV, WR Medical Electronics Co., Stillwater, Minnesota USA). The VM test was standardized to 40 mmHg during 15 s, and the DB test was standardized to six deep breaths during 1 min. Subjects undergoing testing were fasting or had taken breakfast at least 3 h before without coffee or tea; they had emptied their bladder, and had a 15 min relaxation time previous to starting the first test, as well as another 3 min break time between tests. Subjects were also advised to abstain from alcohol and tobacco during the prior 24 h. We used the 2.5th percentile for parameters derived from the CR-R tests as cutoff point of abnormality (Iglesias et al., 1987). For comparative purposes, subjects were divided in two groups according to the absence (CAN-0) or presence of at least one abnormal test for CAN (CAN-1), according to the results of those three tests. eGFR was derived by the Modification of Diet in Renal Disease (MDRD) four-variable equation (Stevens et al., 2006). Results, expressed


945

as ml/min/1.73 m2, were adjusted to the individual body surface area. A 10 yr risk for coronary heart disease (CHD) and stroke was estimated for each subject by the UKPDS Risk Engine v.2, or Framingham scoring system when the former was not applicable according to its authors’ recommendation, adding 2 or 4 points to the total score if the diabetic patient was a woman or a man, respectively. In this case, DM is taken as a categorical variable that does not take into account the influence of DM duration. Patients were further divided in two groups according to the risk for either CHD or stroke using the median values from the present study (10% for the risk of CHD, and 4% for the risk of stroke, respectively). BP Assessment The systolic BP (SBP), diastolic BP (DBP), and HR of each participant were automatically measured every 20 min between 07:00 and 23:00 h and every 30 min during the night for 48 consecutive hours with a properly calibrated SpaceLabs 90207 device (SpaceLabs Inc., Issaquah, Washington, USA). Participants were instructed to go about their usual activities with minimal restrictions but to follow a similar schedule during the two days of ABPM and to avoid daytime napping. BP series were not considered valid for analysis if more than 30% of the measurements were missing, if data were missing for an interval of .2 h, if data were obtained while the subject had an irregular rest-activity schedule, or if the nighttime sleep period was ,6 h or .12 h during ABPM. Protocolcorrect data series were collected from all 101 subjects, with demographic characteristics described in Table 1. Measurement for 48 h was used because it has been previously demonstrated that reproducibility of mean BP values, and thus dipping classification, depends more on measurement duration than measurement frequency (Hermida et al., 2007c). Additionally, just before starting ABPM, we obtained six clinic BP measurements after the subject had rested in a seated position for at least 10 min, using a validated automatic oscillometric device (HEM-705IT, Omron Health Care Inc., Vernon Hills, Illinois, USA). Diagnosis of hypertension was defined as previous treatment for hypertension or office BP 140/ 90 mmHg. Diagnosis in untreated patients was corroborated by ABPM at the time of recruitment. The diagnosis of hypertension based on 48 h ABPM required a diurnal (awake) BP mean 135/85 mmHg and/or a nocturnal (sleep-time) BP mean 120/70 mmHg (Mancia et al., 2007). Actigraphy All participants wore an actigraph (Mini-Motion-Logger, Ambulatory Monitoring Inc., Ardsley, New York, USA) on the dominant wrist to monitor physical activity every min during 48 h ABPM. This compact

946


TABLE 1 Demographic and analytical characteristics of study subjects Variable Patients Sex, % men Smoker, % Type 1 diabetes, % Duration of diabetes, yrs Hypertension, % Antihypertensive treatment, % Microalbuminuria, % Glomerular filtration rate ,60 ml/min/1.73 m2, % Chronic kidney disease, % Age, yrs Height, cm Weight, Kg BMI, Kg/m2 Waist, cm Men with waist 94 cm, % Women with waist 80 cm, % SBP, mmHg DBP, mmHg PP, mmHg HR, beats/min HbA1c, % Creatinine, mg/dL Uric acid, mg/dL Total cholesterol, mg/dL HDL cholesterol, mg/dL LDL cholesterol, mg/dL Triglycerides, mg/dL

101 59.4 22.8 41.6 11.9+7.7 50 37 13.9 7.9 19.8 48.2+11.6 164.5+10.7 78.8+15.7 29.1+4.7 94.7+12.7 72.9 78.6 135.7+17.0 76.4+9.3 59.2+12.0 75.6+9.8 7.8+1.4 0.97+0.18 4.9+1.7 177.7+36.6 46.7+16.0 106.5+32.2 108.8+66.7

Values are mean+SD. Microalbuminuria defined as urinary albumin excretion .30 mg/24 h urine. Glomerular filtration rate estimated using the MDRD-4 equation. Chronic kidney disease defined as glomerular filtration rate ,60 ml/min/1.73 m2 and/or microalbuminuria. Values correspond to the average of six clinic BP measurements obtained per subject in the clinic before commencing 48 h ambulatory BP monitoring. Abbreviations: BMI ¼ body mass index, SBP ¼ systolic BP, DBP ¼ diastolic BP, PP ¼ pulse pressure, HR ¼ heart rate.

device (about half the size of a wristwatch) works as an accelerometer. We synchronized the internal clocks of the actigraph and the ABPM device through their respective interfaces using the same computer. The actigraphy data were used to determine the beginning and end of daytime activity and nocturnal sleep so that the awake and asleep BP means for each patient could be accurately determined.

Statistical Methods Each individual’s clock hour BP and HR values were first referenced to hours after awakening from nocturnal sleep, based on data obtained by


947

wrist actigraphy. This transformation avoided the introduction of bias due to differences among subjects in their sleep/activity routine (Hermida et al., 2002). To correct for measurement errors and outliers, BP and HR were edited according to conventional criteria (Staessen et al., 1991). Thus, readings with SBP .250 or ,70 mmHg, DBP .150 or ,40 mmHg, and pulse pressure (PP, difference between SBP and DBP) .150 or ,20 mmHg were automatically discarded. For descriptive purposes, the circadian rhythm of BP for each group of subjects was objectively assessed by population multiple-component analysis (Fernańdez & Hermida, 1998), a method applicable to nonsinusoidal shaped hybrid time series data (i.e., time series of data collected from a group of subjects, consisting of values distributed at equal or unequal intervals). The circadian rhythm parameters of MESOR (average value of the 24 h rhythmic function fitted to the data) and overall amplitude (onehalf the difference between the maximum and minimum values of the best fitted curve by the method of least-squares) obtained for each group categorized according to CAN and risk of CHD or risk of stroke were compared using a nonparametric test developed to assess differences in parameters derived from population multiple-components analysis (Fernańdez et al., 2004). Hourly BP means obtained for each group, categorized according to CAN and risk of CHD or stroke, were compared by t-test corrected for multiple testing. In so doing, the level of significance was established to be p 0.002 (i.e., the usual level of 0.05 divided by the number of tests done on the same variable—in this case, 24, one for each hourly mean). The awake, asleep, and 24 h BP means as well as the awake/asleep BP ratio were compared among groups using repeated-measures ANOVA. The awake/asleep BP ration is an index of BP dipping, defined as the percent decrease in BP during the hours of nocturnal rest relative to the mean BP obtained during the hours of daytime activity: ðawake BP mean asleep BP meanÞ=ðawake BP meanÞ 100 Apart from the criteria defined above to define hypertension based on ABPM, a 24 h PP mean .53 mmHg was considered abnormal, as this value has been shown to predict adverse prognosis (Verdecchia et al., 1998).

RESULTS Table 1 shows the general characteristics of the sample studied. Among the patients, 45 (44.5%) were non-dipper (rate of 47.6% in type 1 and of 42.4% among those with type 2 diabetes; p ¼ 0.601 between groups),

948


and 50 (50%) patients were hypertensive according to the criteria provided above. Among these, 37 patients were receiving antihypertensive treatment (average 2.2+0.9 medications) at the time of the study, 19 taking all drugs on awakening and the remaining 18 taking at least one drug at bedtime. Fifty percent of the patients had 1 abnormal autonomic test for CAN, and 37% had 2 abnormal tests. Eight patients had eGFR ,60 ml/min/1.73 m2, and 14 had microalbuminuria (albumin excretion rate .30 mg/24 h urine). The total prevalence of chronic kidney disease (eGFR ,60, microalbuminuria, or both) was 19.8% (see Table 1). Figure 1 shows the circadian pattern of SBP (left) and PP (right) measured by 48 h ABPM in the DM participants grouped according to the absence (CAN-0) or presence (CAN-1) of 1 abnormal test for CAN. The dark shading along the lower horizontal axis of the graphs represents the average hours of nocturnal sleep across the subjects. Results did not vary between the two consecutive days of sampling. Therefore, we decided to pool the BP data over an idealized single 24 h profile to simplify the graphic display of the results. SBP was slightly but not significantly (p ¼ 0.124) higher in the CAN-1 subjects. The circadian amplitude of the SBP rhythm was comparable between the two CAN groups (p ¼ 0.348), indicating a similar extent of change in BP throughout the 24 h.

FIGURE 1 Circadian pattern of SBP (left) and PP (right) in DM patients sampled by 48 h ABPM, categorized according to the absence (CAN-0) or presence (CAN-1) of 1 abnormal parameter of CAN. Each graph shows the hourly means and standard errors of data for each group. Dark shading along the lower horizontal axis of the graphs represents the average hours of nocturnal sleep across the sample. The nonsinusoidal shaped curves correspond to the best-fitted waveform model determined by population-multiple-component analysis. MESOR (midline estimating statistic of rhythm) is the 24 h average value of the rhythmic function fitted to the data. Amplitude is one-half the difference between the maximum and the minimum values of the best-fitted curve.

TABLE 2 Ambulatory BP characteristics of study subjects

Variable

949

Patients, n Nocturnal rest, h Awake SBP mean, mmHg Asleep SBP mean, mmHg 24 h SBP mean, mmHg Awake/asleep SBP ratio, % Awake DBP mean, mmHg Asleep DBP mean, mmHg 24 h DBP mean, mmHg Awake/asleep DBP ratio, % Awake PP mean, mmHg Asleep PP mean, mmHg 24 h PP mean, mmHg Awake/asleep PP ratio, % 24 h mean of PP .53, % Non-dipper, %

CAN-0

CAN-1

p between groups

CHD 10%

CHD .10%

p between groups

RFS 4%

RFS .4%

p between groups

51 8.6+1.1 121.3+11.2 106.9+9.1 116.6+9.5 11.6+6.4 75.9+8.1 61.9+6.5 71.4+6.8 18.2+7.3 45.4+6.2 45.0+6.2 45.2+5.9 12.1+4.8 7.8 43.1

50 8.9+1.2 123.9+11.1 111.6+12.2 119.6+10.5 9.8+7.6 71.6+8.6 60.6+6.5 67.8+7.6 14.8+8.1 52.3+9.4 51.0+9.3 51.8+9.0 11.6+5.3 48.0 48.0

0.198 0.253 0.031 0.140 0.193 0.009 0.334 0.013 0.029 ,0.001 ,0.001 ,0.001 0.328 ,0.001 0.624

52 8.8+1.1 119.3+11.2 106.1+9.9 114.9+10.0 11.0+6.1 73.7+9.1 60.8+7.1 69.4+7.9 17.2+7.4 45.6+6.9 45.3+6.7 45.5+6.5 0.2+8.3 9.6 44.2

49 8.8+1.1 126.1+10.1 112.7+11.0 121.6+9.0 10.4+8.0 73.9+8.0 61.8+5.7 69.9+6.9 15.8+8.3 52.2+9.1 50.9+9.1 51.7+8.7 1.8+9.3 38.8 46.9

0.989 0.002 0.002 ,0.001 0.660 0.943 0.411 0.747 0.362 ,0.001 ,0.001 ,0.001 0.350 ,0.001 0.784

53 8.5+1.0 120.9+10.9 107.3+7.8 116.4+8.8 11.0+6.6 75.7+8.4 62.6+5.1 71.4+6.5 16.8+7.2 45.2+5.9 44.7+5.5 45.0+5.5 0.4+8.3 3.7 47.2

48 8.9+1.2 125.0+10.1 113.3+12.6 120.8+10.0 9.4+7.8 70.6+8.4 60.3+6.9 67.0+7.8 14.2+8.0 54.4+9.2 53.0+9.8 53.8+8.9 2.3+9.3 52.1 47.9

0.075 0.083 0.011 0.036 0.332 0.007 0.092 0.006 0.124 ,0.001 ,0.001 ,0.001 0.314 ,0.001 0.975

Values are mean+SD. Patients were categorized according to absence (CAN-0) or presence (CAN-1) of 1 abnormal parameter of cardiac autonomic neuropathy, risk for coronary heart disease (CHD), and risk for stroke (RFS). The awake/asleep ratio (sleep-time relative BP decline), an index of BP dipping, is defined as the % decline in BP during hours of nocturnal rest relative to the mean BP obtained during the hours of diurnal activity, and calculated as: [(awake BP mean – asleep BP mean)/awake BP mean]100. Non-dipper: patients with awake/asleep SBP ratio ,10%, using all data sampled by ABPM for 48 consecutive hours.

950


Table 2 (left columns) compares the awake, asleep, and 24 h means of SBP, DBP, PP, and awake/asleep BP ratio between the subjects of the two groups. Despite a slight increase in asleep SBP in the CAN-1 group, the awake/asleep SBP ratio was comparable between the two CAN groups. DBP was significantly lower among the CAN-1 subjects, mainly during the hours of daytime activity (see Table 2). The prevalence of nondipping was similar in participants with and without CAN (p ¼ 0.624), independent of the type of DM (p ¼ 0.105 for type 1; p ¼ 0.977 for type 2 DM). Among the 24 subjects with CAN-1 and non-dipper BP profile, 12 were untreated, 8 were treated with all drugs on awakening, and 4 were taking at least one drug at bedtime. Among the 26 subjects with CAN-1 and a dipper profile, 11 were untreated, 5 were treated with all drugs on awakening, and 10 were taking at least one drug at bedtime. The tendency to display a more dipper profile associated with bedtime antihypertensive medication treatment was not statistically significant (p ¼ 0.085), probably due to the small sample size. The prevalence of chronic kidney disease was comparable in both CAN groups (p ¼ 0.187), and the eGFR and prevalence of microalbuminuria were also comparable (p ¼ 0.249). In a multivariate analysis, HbA1C % (p ¼ 0.01), DM duration (p ¼ 0.053), and interaction between HbA1C and DM duration (p ¼ 0.04) were the only potential influential factors for non-dipping. The slight elevation in SBP and decrease in DBP in CAN-1 compared to CAN-0 subjects led to a significant difference in PP between the groups (p , 0.001; see Figure 1, right). The elevation of PP in CAN-1 was statistically significant (after correction for multiple testing) for all of the 24 hourly intervals used for descriptive purposes, as indicated by the asterisks above the lower horizontal axis in the right panel of Figure 1. Most relevant, the percentage of participants with an abnormal 24 h PP mean (.53 mmHg) was significantly (p , 0.001) lower in CAN-0 (7.8%) than CAN-1 (48%) subjects (see Table 2). The prevalence of chronic kidney disease was comparable in both the CAN-0 and CAN-1 groups. Figure 2 illustrates the circadian pattern of SBP and PP in diabetic subjects categorized according to their 10 yr risk for CHD, using the median value from this study (risk ¼ 10) as a cutoff threshold. Results shown in Figure 2, and complemented by those shown in Table 2 (center columns), reveal a significant elevation in SBP but not DBP in diabetics with higher CHD risk (p , 0.001). The awake/asleep BP ratio and the prevalence of non-dipping, however, were fully comparable between groups (see Table 2). PP was significantly elevated throughout the 24 h of the day in those diabetic patients with a 10 yr risk for CHD .10 (p , 0.001; see Figure 2, right). When the diabetic participants were categorized according to their 10 yr risk for stroke, the results revealed a significant elevation of SBP and a significant decrease in DBP in those with a risk .4, the median value in

FIGURE 2 Circadian pattern of SBP (left) and PP (right) in DM patients sampled by 48 h ABPM, categorized according to the risk of CHD. Each graph shows the hourly means and standard errors of data for each group. Dark shading along the lower horizontal axis of the graphs represents the average hours of nocturnal sleep across the sample. The nonsinusoidal shaped curves correspond to the best-fitted waveform model determined by population-multiple-component analysis. MESOR (midline estimating statistic of rhythm) is the 24 h average value of the rhythmic function fitted to the data. Amplitude is one-half the difference between the maximum and the minimum values of the best-fitted curve.

FIGURE 3 Circadian pattern of SBP (left) and PP (right) in DM patients sampled by 48 h ABPM, categorized according to the risk for stroke. Each graph shows the hourly means and standard errors of data for each group. Dark shading along the lower horizontal axis of the graphs represents the average hours of nocturnal sleep across the sample. The nonsinusoidal shaped curves correspond to the best-fitted waveform model determined by population-multiple-component analysis. MESOR (midline estimating statistic of rhythm) is the 24 h average value of the rhythmic function fitted to the data. Amplitude is one-half the difference between the maximum and the minimum values of the best-fitted curve.

951

952


this study (see Figure 3 and Table 2, right columns). The circadian BP amplitude, the awake/asleep BP ratio, and the prevalence of nondipping (see Table 2) were comparable between the groups divided according to their 10 yr risk of stroke. The greatest difference between groups was again found for ambulatory PP. The 24 h PP mean was significantly higher, by 8.8 mmHg, in subjects with higher risk for stroke. The percentage of patients with a 24 h PP mean .53 mmHg was significantly higher (p , 0.001) in those with risk for stroke .4 (52%) as compared to those with lower risk (3.7%; see Table 2). The differences in 24 h PP mean between groups categorized according to CAN, risk of CHD, or risk of stroke remained statistically significant after correcting for age. As expected, there was a slight tendency of higher prevalence of chronic kidney disease among subjects with high risk for either CHD or stroke, but the differences in prevalence among the groups categorized according to risk were not statistically significant (p ¼ 0.124). DISCUSSION This study confirms, first, the data previously published on the high coincidence between DM, overweight/obesity (76% of the subjects; 75% had waist 94 [males] or 80 cm [females]; see Table 1), hypertension (50%), and other classical cardiovascular risk factors. Metabolic syndrome (IDF diagnostic criteria) was present in 53% of the participants. Our study shows a prevalence of renal disease, either measured as impaired eGFR (8%), microalbuminuria (14%), or the presence of any of those two conditions (20%), somehow higher than the most recent estimation (11%) in the Spanish general population .20 yrs of age (Otero et al., 2005). Noteworthy is the high prevalence of CAN found in this study: 50% of the whole sample was CAN-1, and 37% CAN 2. Some authors (Hornung et al., 1989; Spallone et al., 1993) have related non-dipping to CAN, suggesting the nocturnal sympathetic predominance relative to parasympathetic activity is the cause of the altered BP pattern. More recently, Spallone et al. (2001) provided data supporting CAN as the pivotal factor of a blunted nocturnal BP fall in both type 1 and type 2 DM, mainly in the latter. These conclusions could not be confirmed in our study, as our results indicate that the prevalence of non-dipping is comparable between patients with and without CAN (see Table 2), independent of the type of DM and the presence or absence of antihypertensive treatment. One of the major differences between the studies is that Spallone et al. (2001, 2007) defined a non-dipper patient using an awake/asleep BP ratio ,0%, which actually refers to a rising BP pattern according to the most commonly used definitions, a much lower prevalent condition than non-dipping (awake/asleep BP ratio ,10%). Other differences are the use by Spallone et al. (2001, 2007) of less reproducible 24 h ABPM and


953

the lack of actigraphy to accurately define the rest and activity period of each individual. In our study, although the asleep but not the awake SBP mean was slightly higher in the CAN-1 subjects, the awake/asleep SBP ratio and prevalence of non-dipping were not significantly different between the CAN groups (see Table 2). Thus, even if CAN might be a factor somehow involved in the genesis of the non-dipper BP pattern in DM, it cannot be the main causal factor. Another recent cross-sectional study (Cardoso et al., 2008) found an increased prevalence of BP non-dipping among patients with type 2 DM and CAN, defined there according to the response to five tests, including the three used in our study plus two additional ones of BP variation during handgrip and standing. Due to the high variability of results, these two additional tests have been discarded more than 15 yrs ago in the evaluation of CAN (Cotzias & Marshall, 1993). Inclusion of these two tests provides a higher than expected proportion of persons showing at least one abnormal test for CAN, which could potentially alter any conclusion regarding the proper relation between CAN and BP non-dipping. Despite these methodological differences, Cardoso et al. (2008) found a significant increase in PP associated with CAN, which is consistent with our present results. The results of our study confirm high awake, asleep, and 24 h PP mean levels in CAN-1 subjects (see Figure 1 and Table 2). PP was also significantly higher in subjects with higher 10 yr risk for either CHD (see Figure 2) or stroke (see Figure 3). A clinically elevated PP level emerges as a main causal factor for the high prevalence of stroke, its recurrence, and mortality in DM (Kissela et al., 2005; Kothari et al., 2002; Tuomilehto et al., 1996). Although in the general population and in subjects with hypertension, controversy exists as to whether or not PP is a cardiovascular risk factor independent from SBP, the supportive evidence for subjects .50 yrs of age is increasing (Franklin et al., 2001). The role of PP in DM has so far received scant attention. The Hoorn Study (Schram et al., 2002) provided first evidence that PP was independently associated with cardiovascular mortality in type 2 DM. More recently, Cockcroft et al. (2005) documented, in a large sample of patients with type 2 DM, that PP was the best predictor of CHD, while SBP was the best predictor for stroke and peripheral vascular disease. These two studies were restricted to type 2 DM, and results were based solely on conventional clinic BP. Our study included patients with both type 1 and type 2 DM, and BP values were obtained by 48 h ABPM. These studies point out the importance of PP as a surrogate marker of aortic stiffness, as it is an independent predictor in DM of mortality and glucose intolerance (Cruickshank et al., 2002) and predictor in older subjects of cardiovascular morbidity and mortality. As indicated by the results of the present study, the prognostic value of PP in DM can be found at a relatively young age (Nakano et al., 2005) as an expression of accelerated atherosclerosis.

954


Despite the lack of statistically significant differences in the prevalence of a non-dipper BP profile between the CAN groups, the results of this study indicate that almost half of the subjects presented a blunted nocturnal decline in BP. The high prevalence of an altered BP pattern in DM might be clinically relevant, as non-dipping has been related to an increase in endorgan injury and cardiovascular events (Boggia et al., 2007; Brotman et al., 2008; Dolan et al., 2005; Ohkubo et al., 2002; Staessen et al., 1999; Sturrock et al., 2000; Verdecchia et al., 1994). Accordingly, there is growing interest as how to best tailor the treatment of hypertensive patients according to their circadian BP pattern (Hermida et al., 2007a; Ohkubo et al., 2002). A number of publications reviewed elsewhere (Hermida et al., 2007a) have documented morning-evening dosing-time differences in the pharmacokinetics and/or pharmacodynamics of several different classes of BP-lowering medications, including angiotensin receptor blockers (Hermida et al., 2007b, 2009), calcium channel blockers (Hermida et al., 2007d; White et al., 1999), angiotensin converting enzyme inhibitors (Kohno et al., 2000; Kuroda et al., 2004; Ohmori & Fujimura, 2005), ablockers (Hermida et al., 2004), ß-blockers (Koga et al., 2005), and diuretics (Hermida et al., 2008b). The results of the present study indicate, in the small subgroup of subjects investigated under the influence of antihypertensive treatment, that there was a borderline significant tendency toward a higher awake/asleep BP ratio among those persons ingesting at least one BP-lowering medication at bedtime, in comparison to those ingesting all medications as a single morning dose upon awakening. The available scientific evidence (Hermida et al., 2007a, 2008a; Portaluppi et al., 1995) suggests that non-dipper patients may benefit from an evening (as opposed to morning) treatment schedule with certain BP-lowering medications to best reduce abnormally high sleep-time BP and to convert the altered non-dipping BP profile to a normal dipper one, which is known to be associated with reduced cardiovascular risk. Whether this approach provides cardiovascular risk reduction and prolongs survival in DM awaits ongoing prospective investigation (Hermida, 2007). In conclusion, the mean HbA1C % and diabetes duration are the unique influential factors of the non-dipper BP pattern found in this sample of DM patients. Cardiac autonomic dysfunction is discarded as a main causal factor of the non-dipper BP pattern in DM. Ambulatory PP, independent of age, is significantly higher in DM patients with cardiac autonomic dysfunction as well as in those with an elevated 10 yr risk for CHD and, particularly, for stroke. DECLARATION OF INTEREST The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.


955

REFERENCES Boggia J, Li Y, Thijs L, Hansen TW, Kikuya M, Bjo ¨rklund-Bodega˚rd K, Richart T, Ohkubo T, Kuznetsova T, Torp-Pedersen CH, Lind L, Ibsen H, Imai Y, Wang J, Sandoya E, O’Brien E, Staessen JA. (2007). Prognostic accuracy of day versus night ambulatory blood pressure: A cohort study. Lancet 370:1219–1229. Brotman DJ, Davidson MB, Boumitri M, Vidt DG. (2008). Impaired diurnal blood pressure variation and all-cause mortality. Am. J. Hypertens. 21:92–97. Cardoso CR, Leite NC, Freitas L, Dias SB, Muxfeld ES, Salles GF. (2008). Pattern of 24-hour ambulatory blood pressure monitoring in type 2 diabetic patients with cardiovascular dysautonomy. Hypertens. Res. 31:865–872. Chau NP, Bauduceau B, Chanudet X, Larroque P, Gautier D. (1994). Ambulatory blood pressure in diabetic subjects. Am. J. Hypertens. 7:487–491. Cockcroft JR, Wilkinson IB, Evans M, McEvan P, Peters JR, Davies P, Scanlon MF, Currie CJ. (2005). Pulse pressure predicts cardiovascular risk in patients with type 2 diabetes mellitus. Am. J. Hypertens. 18:1463–1467. Cotzias C, Marshall JM. (1993). Vascular and electromyographic responses evoked in forearm muscle by isometric contraction of the contralateral forearm. Clin. Auton. Res. 3:21–30. Cruickshank K, Riste L, Anderson SG, Wright JS, Dunn G, Gosling RG. (2002). Aortic pulse-wave velocity ad its relationship to mortality in diabetes and glucose intolerance: An integrated index of vascular function? Circulation 106:2085–2090. Dolan E, Stanton A, Thijs L, Hinedi K, Atkins N, McClory S, Den Hond E, McCormack P, Staessen JA, O’Brien E. (2005). Superiority of ambulatory over clinic blood pressure measurement in predicting mortality: The Dublin outcome study. Hypertension 46:156–161. Ewing DJ, Campbell IW, Clarke BF. (1980). The natural history of diabetic autonomic neuropathy. Quaterly J. Med. 193:95–108. Fernańdez JR, Hermida RC. (1998). Inferential statistical method for analysis of nonsinusoidal hybrid time series with unequidistant observations. Chronobiol. Int. 15:191–204. ń A, Hermida RC. (2004). Comparison of parameters from rhythmometric models Fernańdez JR, Mojo with multiple components on hybrid data. Chronobiol. Int. 21:467–482. Fogari R, Zoppi A, Malamani GD, Lazzari P, Destro M, Corradi L. (1993). Ambulatory blood pressure monitoring in normotensive and hypertensive type 2 diabetes. Prevalence of impaired diurnal blood pressure patterns. Am. J. Hypertens. 6:1–7. Franklin SS, Larson MG, Khan SA, Wong ND, Leip Ep, Levy D. (2001). Does the relation of blood pressure to coronary heart disease risk change with aging: the Framingham Heart Study. Circulation 103:1245–1249. Hermida RC. (2007). Ambulatory blood pressure monitoring in the prediction of cardiovascular events and effects of chronotherapy: Rationale and design of the MAPEC study. Chronobiol. Int. 24:749–775. ń A, Alonso I, Calvo C. (2002). Modeling the circadian Hermida RC, Ayala DE, Fernańdez JR, Mojo variability of ambulatorily monitored blood pressure by multiple-component analysis. Chronobiol. Int. 19:461–481. ´pez JE. Hermida RC, Calvo C, Ayala DE, Domıńguez MJ, Covelo M, Fernańdez JR, Fontao MJ, Lo (2004). Administration-time-dependent effects of doxazosin GITS on ambulatory blood pressure of hypertensive subjects. Chronobiol. Int. 21:277–296. Hermida RC, Ayala DE, Calvo C, Portaluppi F, Smolensky MH. (2007a). Chronotherapy of hypertension: Administration-time dependent effects of treatment on the circadian pattern of blood pressure. Adv. Drug Deliv. Rev. 59:923–939. Hermida RC, Ayala DE, Fernańdez JR, Calvo C. (2007b). Comparison of the efficacy of morning versus evening administration of telmisartan in essential hypertension. Hypertension 50:715–722. ´ n A, Calvo C. (2007c). Influence of measurement Hermida RC, Ayala DE, Fernańdez JR, Mojo duration and frequency on ambulatory blood pressure monitoring. Rev. Esp. Cardiol. 60:131–138. ´pez JE, Rodrı´guez M, Chayań L, Mojo ń A, Fontao MJ, Fernańdez Hermida RC, Calvo C, Ayala DE, Lo JR. (2007d). Dose and administration-time-dependent effects of nifedipine GITS on ambulatory blood pressure in hypertensive subjects. Chronobiol. Int. 24:471–493.

956


Hermida RC, Ayala DE, Fernańdez JR, Calvo C. (2008a). Chronotherapy improves blood pressure control and reverts the nondipper pattern in patients with resistant hypertension. Hypertension 51:69–76. ń A, Chayań L, Domıńguez MJ, Fontao MJ, Soler R, Alonso I, Fernańdez Hermida RC, Ayala DE, Mojo JR. (2008b). Comparison of the effects on ambulatory blood pressure of awakening versus bedtime administration of torasemide in essential hypertension. Chronobiol. Int. 25:950–970. ń A, Fernańdez JR. (2009). Administration-time-dependent Hermida RC, Ayala DE, Chayań L, Mojo effects of olmesartan on the ambulatory blood pressure of essential hypertension patients. Chronobiol. Int. 26:61–79. Hornung RS, Mahler RF, Raftery EB. (1989). Ambulatory blood pressure and heart rate in diabetic patients: An assessment of autonomic function. Diabet. Med. 6:579–585. Iglesias M, Arauxo D, Lado J, Cabezas-Cerrato J. (1987). Computerized simultaneous determination of natural age groups for 32 baroreflex arc function parameters and of the corresponding normal ranges for use in the diagnosis of cardiovascular autonomic neuropathy. Diabet. Res. Clin. Pract. 3:167–173. Ikeda T, Matsubara T, Sato Y, Sakamoto N. (1993). Circadian blood pressure variation in diabetic patients with autonomic neuropathy. J. Hypertens. 11:581–587. Kissela BM, Khoury J, Kleindorfer D, Woo D, Schneider A, Alwell K, Miller R, Ewing I, Moomaw CJ, Szaflarski JP, Gebel J, Shukla R, Broderick JP. (2005). Epidemiology of ischemic stroke in patients with diabetes: The Greater Cincinnati/Northern Kentucky Stroke Study. Diabetes Care 28:355–359. Koga H, Hayashi J, Yamamoto M, Kitamoto K. (2005). Prevention of morning surge of hypertension by the evening administration of carvedilol. Jpn. Med. Assoc. J. 48:398–403. Kohno I, Ijiri H, Takusagawa M, Yin DF, Sano S, Ishihara T, Sawanobori T, Komori S, Tamura K. (2000). Effect of imidapril in dipper and nondipper hypertensive patients: Comparison between morning and evening administration. Chronobiol. Int. 17:209–219. Kothari V, Stevens RJ, Adler AI, Stratton IM, Manley SE, Neil HA, Holman RR. (2002). UKPDS 60: Risk of stroke in type 2 diabetes estimated by the UK Prospective Diabetes Study risk engine. Stroke 33:1776–1781. Kuroda T, Kario K, Hoshide S, Hashimoto T, Nomura Y, Saito Y, Mito H, Shimada K. (2004). Effects of bedtime vs. morning administration of the long-acting lipophilic angiotensin-converting enzyme inhibitor trandolapril on morning blood pressure in hypertensive patients. Hypertens. Res. 27:15–20. Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, Grassi G, Heagerty AM, Kjeldsen SE, Laurent S, Narkiewicz K, Ruilope L, Rynkiewicz A, Schmieder RE, Struijker HA.J, Zanchetti A. (2007). 2007 guidelines for the management of arterial hypertension.The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J. Hypertens. 25:1105–1187. Maser RE, Mitchell BD, Vinik AI, Freeman R. (2003). The association between cardiovascular autonomic neuropathy and mortality in individuals with diabetes: A meta-analysis. Diabetes Care 26:1895–1901. ´brega JC, Cezarini PR, Ferreira SR, Kohlmann O Jr, Ribeiro AB, Zanella MT. Monteagudo PT, No (1996). Altered blood pressure profile, autonomic neuropathy and nephropathy in insulindependent diabetic patients. Eur. J. Endocrinol. 135:683–688. Nakano S, Konishi K, Fukuya K, Uheara K, Nishizawa M, Nakagawa A, Kigoshi T, Uchida K. (2005). A prognostic role of mean 24-h pulse pressure level for cardiovascular events in type 2 diabetic subjects under 60 years of age. Diabetes Care 28:102–107. Ohkubo T, Hozawa A, Yamaguchi J, Kikuya M, Ohmori K, Michimata M, Matsubara M, Hashimoto J, Hoshi H, Araki T, Tsuji I, Satoh H, Hisamichi S, Imai Y. (2002). Prognostic significance of the nocturnal decline in blood pressure in individuals with and without high 24-h blood pressure: The Ohasama study. J. Hypertens. 20:2183–2189. Ohmori M, Fujimura A. (2005). ACE inhibitors and chronotherapy. Clin. Exp. Hypertens. 2 & 3:179–185. Otero A, Gayoso P, Garcıá F, de Francisco AL. (2005). Epidemiology of chronic renal disease in the Galician general population: Results of the pilot Spanish EPIRCE study. Kidney Int. 68(Suppl.):S16–S19.


957

Portaluppi F, Smolensky MH. (1996). Time-dependent and control of arterial blood pressure. Ann. NY Acad. Sci. 783:1–342. Portaluppi F, Vergnani L, Manfredini R, degli Uberti EC, Fersini C. (1995). Time-dependent effect of isradipine on the nocturnal hypertension in chronic renal failure. Am. J. Hypertens. 8:719–726. Portaluppi F, Manfredini R, Fersini C. (1999). From a static to a dynamic concept of risk: The circadian epidemiology of cardiovascular events. Chronobiol. Int. 16:33–49. Portaluppi F, Touitou Y, Smolensky MH. (2008). Ethical and methodological standards for laboratory and medical biological rhythm research. Chronobiol. Int. 25:999–1016. Rutter MK, McComb JM, Forster J, Brady S, Marshall SM. (2000). Increased left ventricular mass index and nocturnal systolic blood pressure in patients with Type 2 diabetes mellitus and microalbuminuria. Diabet. Med. 17:321–325. Schram MT, Kostense PJ, van Dijk RA, Dekker JM, Nijpels G, Bouter LM, Heine RJ, Stehouwer CD. (2002). Pulse pressure and cardiovascular mortality: The Hoorn Study. J. Hypertens. 20:1743–1751. Spallone V, Bernardi L, Ricordi L, Solda` P, Maiello MR, Calciati A, Gambardella S, Fratino P, Menzinger G. (1993). Relationship between the circadian rhythms of blood pressure and sympathovagal balance in diabetic autonomic neuropathy. Diabetes 42:1745–1752. Spallone V, Maiello MR, Cicconetti E, Pannone A, Barini A, Gambardella S, Menzinger G. (2001). Factors determining the 24-h blood pressure profile in normotensive patients with type 1 and type 2 diabetes. J. Hum. Hypertens. 15:239–246. Spallone V, Maiello MR, Morganti R, Mandica S, Frajese G. (2007). Usefulness of ambulatory blood pressure monitoring in predicting the presence of autonomic neuropathy in type I diabetic patients. J. Hum. Hypertens. 21:381–386. Staessen J, Fagard R, Lijnen P, Thijs L, Vaa Hoof R, Amery A. (1991). Ambulatory blood pressure monitoring in clinical trials. J. Hypertens. 9(Suppl. 1):S13–S19. ´ Brien ET, Clement D, de Leeuw PW, Mancia G, Nachev C, Palatini P, Staessen JA, Thijs L, Fagard R, O Parati G, Tuomilehto J, Webster J. for the Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. (1999). Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension. JAMA 282:539–546. Stevens LA, Coresh J, Feldman HI, Greene T, Levey AS. (2006). Assessing kidney function. Measured and estimated glomerular filtration rate. N. Engl. J. Med. 354:2473–2483. Sturrock ND.C, George E, Pound N, Stevenson J, Peck GM, Sowter H. (2000). Non-dipping circadian blood pressure and renal impairment are associated with increased mortality in diabetes mellitus. Diabet. Med. 17:360–364. Tuomilehto J, Rastenyte D, Jousilahti P, Sarti C, Vartiainen E. (1996). Diabetes mellitus as a risk factor for death from stroke: Prospective study of the middle-aged Finnish population. Stroke 27:210–215. Verdecchia P, Porcellati C, Schillaci G, Borgioni C, Ciucci A, Battistell M, Guerrieri M, Gatteschi C, Zampi I, Santucci A, Santucci C, Reboldi G. (1994). Ambulatory blood pressure: An independent predictor of prognosis in essential hypertension. Hypertension 24:793–801. Verdecchia P, Schillaci G, Borgioni C, Ciucci A, Pede S, Porcellati C. (1998). Ambulatory pulse pressure: A potent predictor of total cardiovascular risk in hypertension. Hypertension 32:983–988. White WB, Mansoor GA, Pickering TG, Vidt DG, Hutchinson HG, Johnson RB, Noveck R. (1999). Differential effects of morning and evening dosing of nisoldipine ER on circadian blood pressure and heart rate. Am. J. Hypertens. 12:806–814. Wiegmann TB, Herron KG, Chonko AM, MacDougall ML, Moore WV. (1990). Recognition of hypertension and abnormal blood pressure burden with ambulatory blood pressure recordings in type I diabetes mellitus. Diabetes 39:1556–1560.