International Journal of Cardiology 83 (2002) 133142 www.elsevier.

com / locate / ijcard

Myocardial ischemia and autonomic activity in dippers and non-dippers with coronary artery disease: assessment of normotensive and hypertensive patients
~ ~ Malgorzata Kurpesa*, Ewa Trzos, Jaroslaw Drozdz, Zbigniew Bednarkiewicz, Maria Krzeminska-Pakula
Department of Cardiology, Medical University of Lodz, Bieganski Hospital, Kniaziewicza 1 /5, 91 -347 Lodz, Poland Received 25 September 2001; received in revised form 8 January 2002; accepted 23 January 2002

Abstract Objectives: The aim of this study was to assess the relations between the circadian variations of blood pressure (BP) and the pattern of ischemia and autonomic activity in normotensive and hypertensive patients with coronary artery disease (CAD). Patients and methods: On the basis of the results of ambulatory BP monitoring, 115 patients with stable CAD were divided into Group 1 (with arterial hypertension) and Group 2 (normotensives). Groups were subdivided into dippers and non-dippers. Holter monitoring was performed to assess the occurrence and circadian pattern of ischemic episodes. Time domain and frequency domain HRV analyses were performed to evaluate the autonomic activity. Results: The total number of ischemic episodes was similar in dippers and non-dippers. Non-dippers had a greater number of silent episodes and a different circadian pattern of ischemia with more night episodes. Among the time-domain HRV parameters, only SDNN was similar in dippers and non-dippers. Non-dippers had lower pNN50 and rMSSDthe parameters expressing parasympathetic activity. Differences between diurnal and nocturnal results of spectral HRV analysis were observed in dipper patients only. They presented an elevation of HF power and a decline of LF power at night. All differences between dippers and non-dippers were of similar signicance in both hypertensives and normotensives. Conclusions: A lack of a nocturnal fall in BP is present in normotensive and hypertensive patients with CAD. Non-dippers with CAD had silent and nighttime ischemia more often. They also had an abnormal pattern of autonomic activity with higher sympathetic and lower parasympathetic modulation. 2002 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Dippers; Non-dippers; Coronary artery disease; Myocardial ischemia; Heart rate variability; Autonomic nervous system

1. Introduction Early research using invasive blood pressure (BP) monitoring discovered the characteristic circadian pattern of BP [1]. The introduction of non-invasive automatic blood pressure monitoring (ABPM) allowed study of the circadian rhythm of BP. During

*Corresponding author. Tel. / fax: 148-42-653-9909. E-mail address: kurpesa@ptkardio.pl (M. Kurpesa).

the last decade, non-invasive ABPM has become an established clinical tool [1,2]. The circadian variation of BP is characterized by a diurnal elevation and a nocturnal decline. In several pathophysiological conditions, however, this nocturnal decline is diminished. Thus, hypertensive patients are often subdivided into dippers and non-dippers according to whether they have a greater or smaller decrease in nighttime BP [3]. Today, we know that non-dipping hypertension is not a benign phenomenon. Many studies concerning arterial hy-

0167-5273 / 02 / $ see front matter 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 02 )00031-1

134

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

pertension have demonstrated more serious targetorgan damage in non-dippers [3,4]. The lack of a nocturnal fall in BP is also present in normotensives. However, the signicance of this phenomenon in people without arterial hypertension has not been established. The autonomic nervous system inuences the circadian rhythm of BP [5,6]. The autonomic activity can be assessed using a simple, non-invasive method: the analysis of heart rate variability (HRV). The usefulness of HRV in coronary artery disease (CAD) has been well established. However, the signicance of HRV in arterial hypertension is still not clear. Recent studies have suggested a relationship between the circadian pattern of BP and HRV [7]. The connection between HRV and the circadian pattern of BP in normotensive people still remains unclear. Some authors have documented a higher cardiovascular morbidity in hypertensive non-dippers than in dippers [8,9]. Several studies, using 24-h Holter monitoring, revealed the circadian variability of transient ischemic episodes. However, little is known about the association between the circadian variability of ischemia and the circadian variability of BP. Ischemia and autonomic activity have been established as prognostic determinants in CAD. We hypothesized that it would be useful (for example, for prognostic reasons) to compare such determinants between patients with CAD who had an abnormal and a normal circadian pattern of BP. The results obtained for normotensive dippers and non-dippers appeared to be of special importance since this population remain insufciently investigated. The aim of our study was to assess whether different circadian BP patterns inuence the occurrence of ischemic episodes and autonomic activity in hypertensive and normotensive patients with CAD.

vular heart disease, or clinical signs of heart failure. All patients had to be at sinus rhythm. Those with serious ventricular arrhythmia (IIIV Lown class) and the presence of concomitant diseases which precluded the assessment of ST segment (e.g. bundle branch block, electrocardiographic evidence of left ventricular hypertrophy, implantable pacemaker, signs of preexcitation) were not included. The study group consisted of 115 patients aged from 38 to 72 (mean 5769) years. All medications except short-acting nitrates were discontinued 24 h before the beginning of the study procedures.

2.1. Ofce BP measurements

Systolic and diastolic blood pressure were measured by the conventional sphygmomanometric method in the sitting position [10]. Three measurements at 5-min intervals were performed and the average systolic and diastolic BP values were calculated for each patient.

2.2. 24 -h automatic BP monitoring

ABPM was performed with a non-invasive recorder Tracker NIBP (Reynolds-Medical) on one day (24 h) of typical activity. BP readings were obtained automatically at 15-min intervals from 6 a.m. to 11 p.m. and at 30-min intervals from 11 p.m. to 6 a.m. Average diastolic and systolic BP over 24 h and average daytime and nighttime BP were evaluated. Awake and asleep periods were determined from the patents diaries. According to the VI JNC Report, arterial hypertension was recognized if the average daytime BP was higher than 135 / 85 mmHg and the average nighttime BP was above 120 / 75 mmHg [11]. Based on the difference between the average BP recorded during the day and at night, patients were divided into two subpopulations: dippers (average daytime BP .10% higher than average nighttime BP) and non-dippers (average nighttime BP fall ,10% or nighttime BP higher than daytime).

2. Methods Patients were selected from those with angiographically documented and clinically stable CAD. Subjects were excluded if they had any of the following: myocardial infarction or revascularization procedure within the previous 3 months, idiopathic cardiomyopathy, secondary arterial hypertension, val-

2.3. Holter monitoring

All patients underwent 24-h Holter monitoring using a three-channel recorder Medilog 4500-3 (Ox-

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

135

ford Medical). The magnetic audiotapes (TDK AD 60) were analysed using the computerized system Medilog Excel-2 (Oxford Medical).

2.4. Coronary angiography

All patients had to have angiographically proven CAD to be included in the study. Coronary angiography was performed using Adantx LC1 (General Electric) equipment via the femoral approach using the Judkins technique. Coronary stenoses were expressed as percent diameter reduction. A .70% reduction was considered to be signicant.

2.3.1. Ischemia analysis Transient ischemic episodes were dened as horizontal or downsloping ST segment depression of at last 1 mm measured 80 ms after the J point, lasting for at least 1 min. Episodes were considered to be separate if the ST segment depression was absent for at least 1 min [12]. Episodes classied by the system as ST segment depression were veried by visual step by step analysis. A transient ischemic episode was dened as silent if no symptoms were present at the time of ST segment depression or immediately after.

2.5. Echocardiography
Echocardiography was performed with Acuson 128 XP/ 10C equipment using a 2.54 MHz transducer. The left ventricular ejection fraction (LVEF) was assessed with Simpsons method [14]. The left ventricular mass index (LVMI) was calculated according to the formula of Deveraux and Reichek [16]. Left ventricular hypertrophy (LVH) was dened as LVMI .131 g / m 2 in men and .100 g / m 2 in women [15].

2.3.2. HRV analysis HRV was analyzed according to Standards of the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology [13]. Time domain measures were derived for each 24-h period: SDNN (standard deviation of all selected RR intervals), rMSSD (square root of the mean of the squared successive differences in RR intervals), and pNN50 (percentage of intervals that are at least 50 ms different from the previous interval). Spectral indexes of HRV were computed by an autoregressive model. HRV analysis was performed using Step and Scan mode which allowed division of the Holter recording into 5-min fragments. The following frequency-domain measures were assessed for each fragment: low frequency (LF) (0.04 to 0.15 Hz), high frequency (HF) (0.15 to 0.40 Hz). LF and HF measures were expressed in ms 2 and also as their natural logs (ln). To obtain results for the day, we calculated the averages for all 5-min fragments of the ecg which was recorded between 6 a.m. and 11 p.m. The average values were expressed as LFd and HFd. Similarly, LFn and HFn were assessed as the average values from analyses of all 5-min fragments of the ecg recorded during the nighttime (11 p.m. to 6 a.m.). The LF / HF ratio was calculated separately for both periods of the 24 h: LFd / HFd for the daytime and LFn / HFn for the nighttime.

2.6. Statistical analysis

The mean value6S.D. was calculated for all variables. All data were analyzed by means of analysis of variance (ANOVA). If ANOVA indicated a signicant result the signicance between groups was assessed using a non-parametric chi-squared test. A probability value of less than 0.05 was considered statistically signicant.

3. Results Seventeen patients were excluded because of an uninterpretable Holter monitoring or ABPM result. Thus, the data obtained from 98 patients were analyzed. All subjects (72 male, 26 female, mean age 5568 years) had angiographically documented CAD. Thirty-eight patients (39%) had a history of myocardial infarction. In 26 patients (27%), increased BP values were previously reported, but only 15 of them had been on hypotensive medication. Fifty-ve patients had higher than normal values of BP recorded with 24-h ABPM. They were classied as Group 1 (hypertensives). The remaining 43 patients constituted Group 2 (normotensives). Groups did not differ signicantly with respect to age and gender. In 26

136

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142 Table 2 Blood pressure in the study groups Group 1 Measurement with spygmomanometer (mmHg) Mean systolic BP 154610 Mean diastolic BP 10066 Results of ABPM (mmHg) Mean systolic BPt Mean diastolic BPt Mean systolic BPd Mean diastolic BPd Mean systolic BPn Mean diastolic BPn 15068 9565 160610 105610 13865 8867 Group 2 142610 9068 125610* 7065* 12265* 7265* 11267* 6068*

patients (47%) from Group 1, higher than normal values of BP were previously detected. Only 15 of them have regularly taken antihypertensive medication (ve, betablockers; eight, betablockers1ACEI; two, calcium channel blockers). The duration of treatment varied from 6 to 24 months (mean 14 months). There were no patients with a history of hypertension among the individuals included in Group 2. A family history of CAD was reported by more patients from Group 1 than from Group 2 [21 (38%) vs. 12 (28%), P,0.05] (Table 1). The majority of patients from both groups had single vessel CAD. There were no patients with a narrowed left main coronary artery. Left ventricular EF was similar in both groups (4767% vs. 45610%, P5NS). Twenty-three patients from Group 1 (41%) and two patients (4%) from Group 2 met echocardiographic criteria for LVH.

BP, blood pressure; t, total for 24 h; d, daytime; n, nighttime. *P,0.01, P,0.05.

3.1. Clinical and ambulatory BP data

The mean systolic BP value obtained for the total 98 patients with clinical measurements using a sphygmomanometer was 147610 mmHg and the mean diastolic BP was 9067 mmHg. Patients from Group 1 had a higher BP, but only systolic BP reached statistical signicance (154610 vs. 142610 mmHg, P,0.05). All average BP values obtained with ABPM for the entire 24 h as well as from the daytime and from the nighttime were signicantly higher in Group 1 (Table 2). A nocturnal decrease in BP was recorded in 36
Table 1 Characteristics of the study groups Parameter Number Male Female Age (years) History of myocardial infarction Hypertension Risk factors Diabetes mellitus Hyperlipidemia Smoking Family history of CAD CAD, coronary artery disease. *P,0.001, P,0.05. Group 1 55 41 (75%) 14 (25%) 5466 21 (38%) 26 (47%) 6 (11%) 18 (32%) 40 (72%) 21 (38%) Group 2 43 31 (72%) 12 (28%) 5565 17 (40%) 0* 5 (12%) 13 (30%) 31 (72%) 12 (28%)

patients from Group 1 (65%) and in 34 (79%) from Group 2. These patients were classied as dippers. Nineteen patients from Group 1 (35%) and nine from Group 2 (21%) were non-dippers. There were signicantly more non-dippers in Group 1 than in Group 2 (P,0.01). No differences between dippers and non-dippers were observed with regard to the mean average daytime BP. Conversely, the BP recorded during the night was by denition signicantly lower in dipper patients (Table 3).

3.2. Myocardial ischemia

A total of 290 transient ischemic episodes were recorded: 172 in Group 1 and 118 in Group 2. A tendency to a more frequent occurrence of ischemia was observed among hypertensive patients, but the difference was not signicant. In both groups, about 60% of episodes were silent (103 in Group 1 and 61 in Group 2). The non-signicant tendency to an increased incidence of ischemic episodes was present
Table 3 Comparison of ABPM results between dippers and non-dippers Results of ABPM (mmHg) Group 1 Non-dippers Mean SBPd Mean DBPd Mean SBPn Mean DBPn 15868 10368 14668 9268 Dippers 156610 10566 12866* 8265* Group 2 Non-dippers 12866 7063 12265 6868 Dippers 12764 6866 10668* 5564*

SBP, systolic blood pressure; DBP, diastolic blood pressure; d, daytime; n, nighttime. *P,0.01, dippers vs. non-dippers.

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

137

in non-dippers from both groups (Group 1, 3.261.6 episodes per non-dipper patient vs. 3.061.0 episodes per dipper patient; Group 2, 2.861.3 episodes per non-dipper patient vs. 2.761.2 episodes per dipper patient). Silent ischemia occurred more often in patients without a nocturnal fall in BP. Such a phenomenon was observed in both hypertensives (2.461.2 silent episodes per non-dipper patient vs. 1.660.8 silent episodes per dipper patient, P,0.01) and normotensives (2.160.9 silent episode per nondipper patient vs. 1.661.1 silent episode per dipper patient, P,0.01) The circadian rhythm of ischemia was compared between dippers and non-dippers from both groups. In patients with a nocturnal fall in BP the majority of transient ischemic episodes was recorded during daily activity (75% in dippers from Group 1 and 73% in dippers from Group 2). Conversely, non-dippers had ischemic episodes signicantly more often during the night. Such an inversion of the circadian pattern of ischemia was detected in normotensive and hypertensive dippers (Table 4).

Table 5 Heart rate variabilitytime domain analysis Group 1 Non-dippers SDNN (ms) pNN5O (%) rMSSD (ms) 7169* 3.761.5* 1763.8 Dippers 7662* 4.761.6* 2569.6 Group 2 Non-dippers 92619 4.461.3 2267.9 Dippers 100627 6.261.6 2869.0

*P,0.05 vs. Group 2; P,0.05 vs. non-dippers.

3.3. HRV analysis

SDNN was similar in dippers and non-dippers from both groups. However, hypertensives had a signicantly lower SDNN than normotensives (7169 and 7662 ms vs. 92619 and 100627 ms for dippers and non-dippers from Group 1 vs. Group 2, respectively). pNN50 and rMSSD were also lower in Group 1, but the differences in rMSSD values did not attain statistical signicance (1763.8 ms for non-dippers from Group 1 vs. 2267.9 ms for non-dippers from
Table 4 Results of Holter monitoringmyocardial ischemia in study groups TIE: Silent TIE: Group 1 172 103 (60%) Non-dippers TIE TIE per patient Silent TIE Silent TIE per patient HR during TIE TIE between 06.00 a.m. and 11:00 p.m. 62 3.261.6 45 2.461.2 7668 32 (50%)

Group 2 and 2569.6 ms for dippers from Group 1 vs. 2869.0 ms for dippers from Group 2). Both pNN50 and rMSSD were signicantly lower in non-dippers (Table 5). The parameters of spectral HRV analysis were different in Group 1 and in Group 2. Hypertensive patients (Group 1) had a higher LF power than normotensives (Group 2). Conversely, the values of HF power were signicantly lower in Group 1. Such differences were observed with reference to the values obtained during the day as well as at night. In Group 1 the daytime LF power was similar in dippers and non-dippers. However, at night, the LF power was signicantly higher in non-dippers (12126200 vs. 8586194 ms 2 , P,0.05). Both daytime and nighttime HF values were higher in dippers (HF during the day: dippers 3056146 ms 2 vs. nondippers 2536134 ms 2 , P,0.05; HF at night: dippers 3856147 ms 2 vs. non-dippers 2856143 ms 2 , P, 0.05). The LF / HF power ratio was higher in nondippers. A higher LF / HF was observed during the day (LFd / HFd 4.961.2 vs. 3.860.9, P,0.05) and at night (LFn / HFn 4.461.1 vs. 2.960.6, P,0.05). In Group 2 the spectral HRV analysis performed during the daytime did not reveal differences between

Group 2 118 73 (61%) Dippers 110 3.061.0 58* 1.610.8* ] 72610 82* (75%)* Non-dippers 25 2.861.3 19 2.160.9 7566 13 (52%) Dippers 93 2.761.2 54* 1.661.1* 7567 68* (73%)*

TIE, transient ischemic episodes; HR, heart rate. *P,0.05, dippers vs. non-dippers.

138

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

Table 6 Heart rate variabilityspectral analysis Group 1 Non-dippers LFd (ms ) Ln HFd (ms 2 ) Ln LFd / HFd LFn (ms 2 ) Ln HFn (ms 2 ) Ln LFn / HFn
2

Group 2 Dippers 11936186* 7.0760.15* 3056146* 5.5660.61* 3.860.9* 8586194* 6.7360.22* 3856147* 5.860.43* 2.960.6*

Non-dippers 10256256 6.960.27 4556195 6.060.55 2.760.7 9506270 6.8160.31 3776179 5.7360.37 2.560.4

Dippers 9806261 6.8560.26 4846149 6.1360.34 2.460.8 639624 6.4160.3 5906180 6.3460.31 1.160.2

12596166* 7.1260.13* 2536134* 5.3560.71* 4.961.2* 12126200* 7.1060.19* 2856143 * 5.5160.59* 4.461.1*

Ln, natural logarithm; d, daytime values; n, nighttime values. *P,0.05 vs. Group 2; P,0.05 vs. non-dippers; P,0.05 vs. n.

dippers and non-dippers. However, at night, the dipper patients had a lower LF (6396211 vs. 9506270 ms 2 ) and a higher HF (5906180 vs. 3776179 ms 2 ) than non-dippers. In both groups, patients without a nocturnal fall in BP had similar HRV values during the day and at night. Conversely, in dippers, a clear circadian pattern of HRV values was present with regard to LF in Group 1 and both LF and HF in Group 2. Hypertensive dippers had LF values higher during the day than at night, but HF was stable for the entire 24 h. Dippers from Group 2 (normotensives) presented a nighttime decrease of LF and an increase of HF values (Table 6).

non-dippers with regard to the results of coronary angiography. The total number of ischemic episodes was similar in dippers and non-dippers. However, a lack of a nocturnal fall in BP was associated with a greater number of silent ischemia. Moreover, non-dippers had an abnormal circadian pattern of ischemia. They had more ischemic episodes at night than during the daytime. All differences between dippers and nondippers with regard to the pattern of ischemia were similar in Group 1 and Group 2 (Fig. 1) Some authors have suggested that nocturnal ischemia could be related to the increase of myocardial oxygen demand (increase of BP and heart rate during rapid eye movement sleep) or to the decrease of coronary ow (increased coronary tone and vasoconstriction) [16,17]. In our group, the heart rate during ischemic episodes was similar in dippers and nondippers. However, it is known that high BP increases the myocardial oxygen demand by elevation of the afterload. It is possible that BP which remains unchanged during 24 h may also increase the oxygen demand of cardiac muscle. The constant afterload can

4. Discussion

4.1. Main ndings

In this study, which was performed in a population of patients with angiographically documented CAD, we found a relationship between the circadian pattern of blood pressure and myocardial ischemia as well as autonomic activity. Such a relationship was present in both hypertensive and normotensive patients with CAD.

4.2. Circadian pattern of BP and ischemia

There were no differences between dippers and

Fig. 1. Circadian pattern of ischemia in study groups. The percentage of TIE recorded during the day and at night in Dippers (part A) and Non-Dippers (part B) is compared between Group 1 (solid bars) and Group 2 (open bars).

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

139

lower the ischemia threshold by a desensitization-like mechanism. Under this condition, even a minor change of BP and heart rate during sleep, which are hardly perceptible in people with a nocturnal fall in BP, could provoke an episode of ischemia in nondippers. The other possible explanation for the frequent nocturnal ischemia in non-dippers is that stable BP levels during the day and night may accelerate the dysfunction of endothelium [18]. This could lead to the domination of vasoconstriction, which usually increases the myocardial oxygen demand. Moreover, coronary vasoconstriction also reduces myocardial perfusion. Therefore, the coexistence of two mechanisms (increased oxygen demand and decreased oxygen supply) seems to be involved in the pathogenesis of the nocturnal intensication of ischemia in non-dippers. There are very few studies on the relation between the circadian pattern of BP and ischemia. Pierdomenico et al. [9] reported a more frequent presence of ischemia during the night in non-dippers. However, they only assessed hypertensive patients. In their study, silent ischemia was similar in dippers and non-dippers. Conversely, we found more silent ischemic episodes in non-dippers. The explanation for such a phenomenon is not easy because the etiology of silent ischemia is still not clear [19]. Previously, it was suggested that an assessment of autonomic activity may at least partly clarify this problem. The autonomic nervous system has been extensively implicated in the mechanism of cardiac pain perception. The afferent sympathetic bers are of crucial importance in the transmission of cardiac pain. Some investigators have also reported the role of afferent parasympathetic bers in the phenomenon of the radiation of cardiac pain. It is known that diabetic patients very often have silent ischemia. Some authors have documented that a lack of a nocturnal fall in BP could accelerate the development of autonomic neuropathy in diabetes mellitus [20,21]. However, we also found an association between a lack of a nocturnal fall in BP and autonomic activity in patients without diabetes. It is possible that disturbances of the circadian pattern of BP are only the symptoms of changes in autonomic activity which could promote the occurrence of silent ischemia not only in diabetic but in all non-dipper patients with coronary artery disease.

4.3. Circadian pattern of BP and autonomic activity

In this study, we have documented the association between the circadian pattern of BP and autonomic activity. Such an association was present in all patients with CAD: normotensives and hypertensives. In the time domain HRV analysis, only SDNN was similar in dippers and non-dippers. The other parameters (pNN50 and rMSSD) were lower in the group without a nocturnal fall in BP. The difference had the same signicance in both normotensives and hypertensives. pNN50 and rMSSD are parameters usually correlated with parasympathetic activity [22]. Our results for the time domain HRV analysis indicated a depression of parasympathetic activity in non-dippers. Similar results were obtained by Abate et al., who evaluated the autonomic activity during the tilt test, deep breathing and coughing [23]. Parasympathetic attenuation is known to be a determinant of an unfavourable prognosis in CAD [24]. Our ndings indicate that non-dippers with CAD are likely to constitute a group at high risk for new coronary events and it concerns not only those with hypertension but also normotensives. Several investigators have postulated the importance of increased sympathetic activity in the pathogenesis of arterial hypertension [25]. Our results for the power spectral HRV analysis are consistent with such opinions, because we found a higher LF power in Group 1. However, we also found a signicantly lower HF power, which was observed during daily activity and at night in this group. Low HF may be an indicator of impairment of parasympathetic activity in hypertensive patients. Recent studies have postulated the role of decreased parasympathetic activity in the etiology of hypertension [26]. Our results seem to conrm these reports. We also found abnormalities of autonomic activity in non-dippers. They had a higher LF power and higher LF / HF ratio at night. Analysis of the results obtained showed an increase of HF power and a decrease of LF power at night in dipper patients only. These results suggest that abnormal autonomic activity may play a role in the etiology of the non-dipping phenomenon and such a mechanism is also important in normotensives. Several studies have reported abnormalities of autonomic activity in non-dippers [2730]. However,

140

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

all these studies evaluated hypertensive patients only. As far as we are aware, this study is the rst report which also includes normotensive patients. We have documented that, in this group, the association between abnormalities of the circadian pattern of BP and HRV is similar as in hypertensives, although it is present only at nighttime. Conversely, in patients with hypertension, all abnormalities of autonomic activity in non-dippers also occurred during the day. The increase in sympathetic activity and parasympathetic depression which we found in non-dippers may explain the greater number of ischemic episodes in this group. Autonomic abnormalities may also play a role in the change of the ischemic pain threshold in non-dippers.

5. Clinical implications Patients with CAD who have transient ischemia in Holter monitoring are known to be a high risk population [31,32]. The presence of silent ischemia is also a determinant of an unfavourable prognosis [33]. We have found that patients with CAD and without a nocturnal fall in BP (non-dippers) often have ischemia at night. They are also likely to present silent ischemia. Moreover, non-dippers presented with such autonomic abnormalities, which are known to be dangerous for patients with CAD, because several investigators have documented a higher mortality among patients with ischemic heart disease and decreased parasympathetic activity [24,33]. It seems that ABPM could be helpful in patients with CAD, not only in the determination of the group with hypertension and a higher risk of target-organ damage, but also in the prediction of new coronary events. This new advantage of ABPM refers to both normotensive and hypertensive patients with ischemic heart disease.

patients had a funduscopic examination. Vascular changes (Grades 1 to 3) were present in all from Group 1 and in only one patient from Group 2. All patients in our study population had to be treated because they suffered from CAD. It is known that some medications can inuence the results of ischemia monitoring or HRV analysis. On the other hand, sudden withdrawal of the treatment may cause exacerbation of the disease. We stopped all medications 24 h before ABPM and 48 h before Holter monitoring. Since none of our patients had been treated with long-acting antiarrhythmic drugs such as Amiodaron we believe that such a wash-out period was sufcient to obtain reliable results and was safe for the patients. The reproducibility of the dipper status seems to be controversial. Many factors may inuence the circadian rhythm of blood pressure, including smoking, and alcohol and sodium intake. Such factors could increase the diurnalnocturnal difference. According to Mancia et al., correct indices of blood pressure variability can only be obtained by beat-to-beat monitoring. Blood pressure monitoring by the usual non-invasive, intermittent devices provides less-reliable indices of variability [34]. In some studies, up to 30% of non-dippers converted to dippers during automatic blood pressure monitoring [35]. However, Zakopoulos et al. found a high reproducibility of all measurements, including intervals limited to 1 h, in a study of serial ABPM performed over a period of 4 months [36]. At present, insufcient clinical data are available to allow a denitive comment on the problem of the reproducibility of circadian BP variation. However, recent studies suggest that ABPM may be a useful tool for blood pressure changes. We focused on the importance of the non-dipping phenomenon in normotensive patients. However, in our opinion, cardiovascular chronobiology is an interesting new area of investigation. Large studies would be necessary to further explore the importance of chronobiology in modern cardiology.

6. Study limitations We divided our population into normotensives and hypertensives according to the results of 24-h ABPM only. Prolonged ABPM were not performed. We did not evaluate the eye fundus before inclusion in the study. However, after termination of the study, all 7. Conclusions 1. A lack of a nocturnal decrease in BP is a common nding in patients with coronary artery disease.

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142

141

2. Patients with coronary artery disease and without a nocturnal fall in BP more often have silent ischemia. 3. Normotensive and hypertensive patients with coronary artery disease and an abnormal circadian pattern of BP also have an abnormal circadian pattern of ischemia and abnormal autonomic activity.

Acknowledgements This study was supported by research grant No. 502-11-500 (128) from the Medical University of Lodz.

References
[1] Pickering TG, OBrien E. Second International Consensus Meeting on Twenty-Four-Hour Ambulatory Blood Pressure Measurement: consensus and conclusions. J Hypertens 1999;9(Suppl 8):S26. [2] Mancia G. Ambulatory blood pressure monitoring: research and clinical implications. J Hypertens 1990;8(Suppl 7):S1S13. [3] OBrien E, Sheridan J, OMalley K. Dippers and non-dippers (letter). Lancet 1988;2:397. [4] Verdecchia P, Schillaci G, Guerrieri M et al. Circadian blood pressure changes and left ventricular hypertrophy in essential hypertension. Circulation 1990;81:52836. [5] Floras JS, Hassan OM, Jones JV. Factors inuencing blood pressure and heart rate variability in hypertensive humans. Hypertension 1988;11:273140. [6] James CG, Pickering TG. The inuence of behavioral factors on the daily variations of blood pressure. Am J Hypertens 1993;6:170S 5S. [7] Huikuri HV, Ylitalo A, Pikkujamsa SM et al. Heart rate variability in systemic hypertension. Am J Cardiol 1996;77:10737. [8] Verdecchia P, Porcellati C, Schillaci G. Ambulatory blood pressure: an independent predictor of prognosis in essential hypertension. Hypertension 1994;24:793801. [9] Pierdomenico SD, Bucci A, Costantini F, Lapenna D, Cuccurullo F, Mezzeti A. Circadian blood pressure changes and myocardial ischemia in hypertensive patients with coronary artery disease. J Am Coll Cardiol 1998;31:162734. [10] Korotkow N. To the question of methods of determining the blood pressure. Rep Imp Milit Acad 1905;11:3657. [11] The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure. Arch Intern Med 1997;157:241345. [12] Cohn P, Kamel W. Recognition, pathogenesis and management options in silent coronary artery disease. Circulation 1987;75(Suppl):I54. [13] Task Force of The European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standards of measurement, physiological interpretation and clinical use. Eur Heart J 1996;17:35481.

[14] Vuille C, Weyman AE. Left ventricle I: general considerations, assessment of chamber size and function. In: Weyman AE, editor, Principles and practice of echocardiography, 2nd ed., Lea & Febiger, 1994, pp. 575624. [15] Devereux RB, Reicheck N. Echocardiographic determination of left ventricular mass in men: anatomic validation of the method. Circulation 1977;55:6138. [16] Yeung AC, Raby KE, Ganz P, Selwyn A. New insights into the management of myocardial ischemia. Am J Cardiol 1992;70:8G 13G. [17] Pepine CJ. New insights in measurement of myocardial ischemia. Am J Cardiol 1992;70:19G25G. [18] Luscher TF, Noll G. Endothelium dysfunction in the coronary circulation. J Cardiovasc Pharmacol 1994;24(Suppl 3):1626. [19] Deedwania P, Nelson JR. Pathophysiology of silent myocardial ischemia during daily life. Circulation 1990;82:1296304. [20] Hikita H, Kurita A, Takase B et al. Usefulness of plasma betaendorphin level, pain threshold and autonomic function in assessing silent ischemia in patients with and without diabetes mellitus. Am J Cardiol 1993;72:1403. [21] Lindsay RS, Stewart MJ, Naim IM, Baird JD, Padeld PL. Reduced diurnal variation of blood pressure in non-insulin-dependent diabetic patients with microalbuminuria. J Hum Hypertens 1995;9:2237. [22] Kleiger RE, Bosner MS, Rottman JN. Time-domain measurements of heart rate variability. J Ambul Monit 1993;6:18. [23] Abate G, DAndrea L, Battestini M, Zito M, Di Iorio A. Autonomic nervous activity in elderly dipper and non-dipper patients with essential hypertension. Aging (Milano) 1997;6:40814. [24] De Ferrari GM, Vanoli E, Schwartz PJ. Cardiac vagal activity, myocardial ischemia and sudden death. In: Zipes DP, Jalife J, editors, Cardiac electrophysiology. From cell to bedside, Philadelphia: W.B. Saunders, 1995, pp. 42234. [25] De Qoattro V, De-Ping Lee D. Blood pressure reactivity and sympathetic hyperactivity. Am J Hypertens 1991;4(Suppl):624S 9S. [26] Langewitz W, Ruddel H, Schachinger H. Reduced parasympathetic cardiac control in patients with hypertension at rest and under mental stress. Am Heart J 1994;127:1228. [27] Frissina N, Pedulla M, Mento G, Lanuzza B. Frequency domain of heart rate and blood pressure variability in essential hypertensive patients during asleep: differences between dippers and non-dippers. Blood Press Monit 1996;1:42531. [28] Vaile JC, Stallard TJ, al-Ani M, Jordan PJ, Townend JN, Littler WA. Sleep and blood pressure: spontaneous baroreex sensitivity in dippers and non-dippers. J Hypertens 1996;14:142732. [29] Pedulla M, Silvestri R, Lasco A et al. Sleep structure in essential hypertensive patients: differences between dippers and non-dippers. Blood Press Monit 1995;4:2327. [30] Hojo Y, Noma S, Ohki T, Nakajima H, Satoh Y. Autonomic nervous system activity in essential hypertension: a comparison between dippers and non-dippers. J Hum Hypertens 1997;10:66571. [31] Kurpesa M, Krzeminska-Pakula M, Bednarkiewicz Z, Trzos E, Peruga JZ, Kosmider M. Ischaemic episodes detected by early exercise test and Holter monitoring after myocardial infarction: which are of most prognostic value? Coron Artery Dis 1996;7:789 96. [32] Hedblad B, Juul-Mooler S, Svensson K et al. Increased mortality in men with ST segment depression during 24-h ambulatory long-term electrocardiogram recording. Eur Heart J 1989;10:14955. [33] Gottlieb SO, Gottlieb SH, Achuff S et al. Silent ischemia on Holter monitoring predicts mortality in high-risk postinfarction patients. J Am Med Assoc 1988;259:103041.

142

M. Kurpesa et al. / International Journal of Cardiology 83 (2002) 133 142 [36] Zakopoulos NA, Vemmos KN, Kotsis VTh, Pitiriga VCh, Stamatelopoulos SF, Moulopoulos SD. Reproducibility of ambulatory blood pressure measurements in essential hypertension. Blood Press Monit 2001;6:415.

[34] Wilson DK, Sica DA, Dereus M, Nicholson SC. The inuence of potassium intake on dipper and non-dipper blood pressure status in an African-American adolescent population. Blood Press Monit 1996;1:44756. [35] Mancia G, Di Renzo M, Parati G. Ambulatory blood pressure monitoring use in hypertension research and clinical practice. Hypertension 1993;21:51024.