Early Morning Surge and Dipping Status of Blood Pressure: Are These of Predictive Value for Silent Myocardial Ischemia?

Abstract

To investigate the predictive value of morning surge (MS) and dipping status of blood pressure (BP) for ST‐segment depression (ST depression) in hypertensive persons, the combined 24‐hour ambulatory BP measurement and Holter electrocardiographic recordings of 344 patients (132 women and 212 men, 60±12 years) were analyzed. The morning BP surge was calculated as mean systolic BP during the 2 hours after awaking minus the mean systolic BP during the 1 hour that included the lowest sleep BP. Depending on the percentage decrease in systolic BP at night, patients were subdivided into extreme dippers, dippers, nondippers, and risers. ST depression was defined as horizontal or descending ST depression of 1 mm of 1 minute duration and a 1‐minute interval after the previous episode or 24‐hour mean ST‐segment value <−0.1 mV. ST depression was observed in 76 (22.1%) of 344 patients. The mean MS in patients with ST depression was with 25±15 mm Hg and not significantly different when compared with patients without ST depression (26±15 mm Hg). The prevalence of ST depression was not significantly different in dippers, extreme dippers, nondippers, and risers (20%, 23.5%, 23%, and 25.5%, respectively). Systolic and diastolic BP values during ST depression were significantly higher in the morning as compared with ST depression in the evening (149/84 mm Hg vs 138/78 mm Hg, P<.05). In the present study, MS and dipping status of BP were not associated with ST depression. ST depression was, however, characterized by significantly higher BP peaks in the early morning hours.

Blood pressure (BP) and heart rate fluctuations have been discussed as hemodynamic mechanisms triggering silent cardiovascular ischemic events. The extent of the morning surge and the day‐night variability of BP are thought to be unfavorably associated with silent cardiovascular events. In the 24‐hour BP analysis, the day‐night adaptation of BP is subdivided into 4 categories: dippers, extreme dippers, nondippers, and risers. For extreme dippers, nondippers, and risers, an especially high risk of cerebrovascular and cardiovascular events is discussed. ¹ However, Kario and colleagues ² have described a stricter association of silent cerebrovascular events with a steep rise of BP in the morning than with a fall in BP at night in elderly hypertensives. In the opinion of some authors, this steep rise in BP in the morning (also termed early morning surge [MS]) will trigger cardiovascular events such as myocardial infarction, sudden cardiac death, and myocardial ischemias to occur preferentially in the early hours of the morning. ³ , ⁴ There is still controversy as to whether the phenomenon of the MS is a manifestation of inadequate antihypertensive therapy or of a pathologically increased adaptation reflex. However, according to most investigators, prognosis is considered to be unfavorable. Another phenomenon observed with a high prevalence in hypertensive patients is ST depression detected by an analysis of the ST‐segment changes on a Holter‐electrocardiographic (ECG) monitor. ⁵ , ⁶ These can be triggered by variations of BP in the early morning and also have an unfavorable cardiovascular prognosis. ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ The objective of the present study in hypertensives was hence to investigate the predictive value of MS and dipping status of BP for ST‐segment depression and whether these parameters affect the circadian rhythmicity of ST‐segment depression episodes.

Methods

Between 1999 and 2004, out‐clinic patients with arterial hypertension were investigated by means of CardioTens monitoring (Meditech, Budapest, Hungary) in Bonn, Germany. The CardioTens recordings from 384 patients (132 women and 212 men, 60±12 years) in which the sleeping and waking phases were defined in the protocols of each patient’s day were analyzed in this study. A blinded analysis of the results was carried out. Treated or untreated arterial hypertension and suspected arterial hypertension were the inclusion criteria. The following characteristics were the exclusion criteria: valvular insufficiencies, pericarditis, cardiomyopathies, traces of heart blocks on the resting ECG, electrolyte imbalances in the serum, pre‐excitation syndrome, taking digitalis, atrial fibrillation, numerous ventricular extrasystoles, or presence of a pacemaker.

Combined 24‐Hour ABPM and Long‐Term ECG Recording

CardioTens is a combined instrument for recording long‐term BP and the ECG. It is able to analyze the ST segment continuously, to store ECGs and to ensure that ABPMs are carried out. Furthermore, the instrument triggers an additional BP measurement during an ST depression which fulfills the criteria that defines an ischemic episode. An intermittent ST depression was defined in accordance with the 1‐1‐1 rule (horizontal or descending ST depression of 1 mm and 1 minute duration— 1‐minute interval after the previous episode). Before the beginning of the 24‐hour investigation, the isoelectric reference point (PQ segment) of the J point and the I point (80 ms after the J point) and an ST‐segment detection interval of at least 3 mm as the initial ST level was calculated individually for each patient. If required, the investigator was able to manually correct this initial ST‐detection interval calculated by the instrument. At an initial ST level of <0 mV, the ST segment had to be additionally lowered by –0.1 mV compared with the initial ST in order to be rated as an intermittent ST‐depression episode. If the ST‐segment analysis revealed a 24‐hour mean value of ≤−0.1 mV, the investigation was appraised as a registration with permanent ST‐segment depression. For methodologic reasons, the ST analysis of these investigations with permanent ST depression was not included in the evaluation of the intermittent ST depressions.

The method of detecting ST‐segment depression with the CardioTens monitoring has been published previously. ¹⁴

The BP parameters investigated included the following: 24‐hour mean values for BP, heart rate, and pressure‐frequency product. In patients with intermittent ST‐segment depression, these parameters were measured during the ST‐segment depression. If present, BP values in a 10‐minute interval before and after an ST‐segment depression episode were also evaluated. In the evaluation of the BP values, the following parameters are analyzed:

• • Awaking period: mean values 2 hours before awaking
• • BP in the morning: mean values 2 hours after awaking
• • Mean values 2 hours before sleeping
• • Mean values for the BP parameters during sleep (nighttime)
• • Mean values for BP in the daytime
• • Lowest BP value: mean value of the lowest systolic BP during the nighttime and the measurements immediately before and after this BP value
• • The morning surge was calculated as follows: BP in the morning − lowest BP value

The extent of the morning surge was compared between patients with and without ST depression. Depending on the percentage decrease in the systolic BP in the passive period in accordance with the formula: (100 × [1−passive period/active period]), the patients were subdivided into 4 groups: extreme dippers ≥20%, dippers <20% but ≥10%, nondippers <10% but ≥0%, and risers <0%. The influence of dipping status on prevalence and circadian distribution of intermittent ST depressions were analyzed.

The patient‐related data such as age, height, body weight, sex, office BPs, duration of hypertension and treatment, type of antihypertensive medication, presence of cardiovascular risk factors (diabetes mellitus, smoking, dyslipoproteinemia, coronary heart disease, myocardial infarction, peripheral arterial occlusion disease, sleep apnea), angina pectoris according to New York Heart Association stages and dyspnea according to Canadian Chest Society stages were recorded. If available data on total cholesterol, low‐density lipoprotein and high‐density lipoprotein fractions, Sokolov index, appraisal of the pump function in cardiac ultrasound, ergometric finding, myocardial scintigraphy and coronary angiography, revascularization measures (percutaneous transluminal coronary angioplasty, stent and lysis) were also evaluated.

Statistic

BPs were analyzed together with the other qualitative and quantitative parameters using SPSS 12.1 software (SPSS Inc, Chicago, IL). The statistical methods used were the mean, standard deviation and, where indicated, Student t test and Fisher Exact test.

Results

The CardioTens records of 344 patients (60±12 years, 242 men and 140 women) were investigated. A total of 299 (87%) patients were being treated with antihypertensive medication; 24 (7%) were untreated. There was no information on antihypertensive treatment in 21 patients. Seventy‐two of the patients had arterial hypertension for more than 11 to 20 years and 117 for 1 to 10 years. In 60 patients, arterial hypertension had been diagnosed <1 year earlier or had been recently diagnosed. In 95 patients, this information was not evaluated. Seventy‐six patients had diabetes mellitus, and 69 were smokers. In 192 patients, dyslipoproteinemia was known. The lipid values were as follows: total cholesterol, 209±50 mg/dL; low‐density lipoprotein cholesterol, 136±47 mg/dL; and high‐density lipoprotein cholesterol, 49±17 mg/dL.

ST depression was observed in 76 (22.1%) of 344 patients (Table I). Patients with ST depression showed significantly lower body weight and body mass index and significantly higher systolic BP and BP amplitudes than patients without ST depression. There were no significant differences between the 2 groups with regard to systolic BP, heart rate, sex, height, smoking, dyslipoproteinemia, lipid values, and type of antihypertensive treatment (β‐blockers, diuretics, calcium channel blockers, angiotensin‐converting enzyme inhibitors, AT₁‐receptor blockers, or the frequency of monotherapy or treatment with 2–5 medications). Intermittent ST depression was observed in 47 patients.

Table I.

 Comparison of Patient Characteristics Between Patients With and Without ST Depression

	With (n=76)	Without (n=268)	Student t Test/χ2 Test
Age, y	60±14	59±12	NS
Men (n=212)	44	168	NS
Weight, kg	80±15	84±16	.036
Height, m	1.71±.08	1.71±.09	NS
Body mass index, kg/m2	27±5	29±5	.021
Mean morning surge, mm Hg	25±15	26±15	NS
Morning surge ≥45 mm Hg (n=34)	8	26	NS
Dipping status
Dipper (n=140)	28	112	NS
Extreme dipper (n=31)	7	24	NS
Nondipper (n=126)	29	97	NS
Riser (n=47)	12	35	NS
Systolic BP values, mm Hg
24‐hour	137±17	124±16	.001
Awaking period	132±22	123±20	.001
2 hours after awaking	140±18	134±20	.035
Daytime	140±17	135±16	.010
2 hours before  sleeping	139±20	132±19	.002
Nighttime	129±19	122±18	.002
Lowest BP value	119±21	112±18	.004
Office BP	149±22	144±21	NS
Diastolic BP values, mm Hg
24‐hour	76±10	74±9	NS
Awaking period	77±13	75±12	NS
2 hours after awaking	69±11	67±10	NS
Daytime	80±11	78±10	NS
2 hours before sleeping	71±13	69±12	NS
Nighttime	79±10	78±11	NS
Lowest BP value	63±11	61±11	NS
Office BP	84±14	85±14	NS
Heart rate, beats per min
24‐hour	73±11	72±11	NS
Awaking period	66±13	64±10	NS
2 hours after awaking	67±12	66±10	NS
Daytime	77±12	76±12	NS
2 hours before sleeping	74±12	72±11	NS
Nighttime	74±13	73±12	NS
Lowest BP value	64±11	64±10	NS
Office BP	75±11	74±13	NS
Pulse pressure, mm Hg
24‐hour	61±14	55±12	.001
Daytime	61±14	55±12	.001
Nighttime	60±15	54±13	.007

Abbreviations: BP, blood pressure; NS, not significant.

In the remaining 29 patients with ST depression, the ST‐segment analysis showed a 24‐hour mean value, which was lower than −0.1 mV. These patients were hence no longer included in the analysis of the intermittent ST depression. The mean frequency of ST depression in the 47 patients with intermittent ST depression was 13±12 episodes, the mean duration was 3.76±3.04 minutes, and the mean ST depression was −0.15±0.04 mV. The circadian distribution of the intermittent ST episodes is shown in Figure 1. On average, ST depression in the period from 6 am to 9 am lasted longer than ST depression during the remaining part of the day (9 am–6 am) (5.8±6.1 vs 3.5±3.4 minutes; P=.057). The mean ST‐segment depression was the same in both daily intervals (−0.15±0.04 vs −0.15±0.04 mV; P=.403). ST depression in the period between 6 am and 9 am were characterized by significantly higher BPs before, during, and after the ST depression as compared with ST depression in the period between 6 pm and 9 pm (Figure 2). Ninety‐four percent of the intermittent ST depression was silent.

Figure 1.

 Circadian distribution of intermittent ST‐segment depressions.

Figure 2.

 Comparison of blood pressure (BP) parameters at the onset of ST depression between 6 am and 9 am (black bars) vs 6 pm and 9 pm (white bars). *t test: P<.05.

The extent of the mean morning surge in patients with ST depression was 25±15 mm Hg, not significantly different when compared with patients without ST depression (26±15 mm Hg).

Thirty‐one (9%) patients were extreme dippers, 140 (41%) were dippers, 126 (37%) were nondippers, and 47 (13%) were risers. The BP and ST changes of dippers, extreme dippers, nondippers, and risers are shown in Table II. It is striking that in all 4 groups, higher BPs than 24‐hour ABPM mean values are measured at the time of ST depression.

Table II.

 Comparison of BP and ST Parameter in Dippers, Extreme Dippers, Nondippers, and Risers With Intermittent ST Depression

Group	Dipper (n=22)	Extreme Dipper (n=5)	Nondipper (n=16)	Riser (n=4)
24‐hour ambulatory BP monitoring
Systolic	130±12	130±10	140±18	140±20
Diastolic	73±10	78±6	77±12	75±10
Heart rate	73±11	79±9	70±9	70±14
Pressure‐frequency product	9401±1406	10,303±1803	9729±1637	9604±1093
ST depressions per patient	13±14	13±6	15±12	11±8
Mean ST duration, min	3.23±2	3.46±1	4.42±4	4.49±3
Mean ST depression, mV	−.13±.03	−014±.05	−.16±.04	−.14±.02
BP parameters at the onset of ST depression
Systolic	142±17a	145±19	152±20	141±24
Diastolic	83±16a	88±15	85±11	79±18
Heart rate	85±15a	92±13	84±16a	76±18
Pressure‐frequency product	11,794±2822a	13,474±3112	12,767±2817a	10,325±1339

^a t test P<.05 of the parameter as compared with the corresponding 24‐hour ambulatory blood pressure (BP) monitoring mean value.

Since the number of cases of extreme dippers and risers was too small, the extreme dippers were grouped together with dippers, and risers with nondippers for the analysis of intermittent ST depression. Intermittent ST depression revealed differences between dippers (+extreme dippers) and nondippers (+risers) (Table III). There was a tendency for the duration of ST depression and the depth of ST depression to be more pronounced in nondippers (+risers) than in dippers (+extreme dippers) although this difference was not statistically significant. Whereas 32% of the ST depression in dippers (+extreme dippers) occurred in the evening between 7 pm and midnight, and only 16% in nondippers (+risers) (Table II). The pressure‐frequency product before, during, and after an ST depression showed characteristics comparable with those of the 24‐hour ABPM mean value in dippers (+extreme dippers) and nondippers (+risers), with a maximum rise at the time of ST depression (Figure 3).

Table III.

 Comparison of ST Parameters Between Dippers (+Extreme Dippers) and Nondippers (+Risers)

ST Depression	Mean Duration, Min	Mean Depression, mV	Per Patient, 24‐Hour	Per Patient, 6 am–9 am	Per Patient, 7 pm–12 pm
Dippers (n=27)	3.28±2.02	−.14±.03	13±13	2.7	4.3
Nondippers (n=20)	4.53±4.01	−.16±.04	14±12	2.6	2.3
t test	NS	NS	NS	NS	P<.05

Abbreviation: NS, not significant.

Figure 3.

 Comparison of pressure‐frequency product before, during, and after ST depression: dippers vs nondippers. Student t test: Pressure‐frequency product before, during, and after ST depression as compared with the corresponding 24‐hour mean value.

Discussion

In recent years, working groups who have investigated the morning surge of BP have suggested that an enhanced morning surge is highly correlated with cardiovascular events. ² , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ Only one study group, however, has succeeded in demonstrating that a raised early morning surge is associated with cardiovascular or cerebrovascular ischemic events. ² The results of this study group have not been confirmed by any other study group.

Kario calculates the morning surge as a difference between the systolic mean values of an interval 2 hours after getting up minus a 1‐hour interval that includes the minimum systolic BP value at night. However, he defines a cut‐off value of >55 mm Hg in one paper, ² >45 mm Hg in a second paper, ¹⁶ and >36 mm Hg in a third paper ¹⁷ as an increased morning surge. A uniform definition for calculating the morning surge and generally accepted cut‐off values to distinguish a normal adaptation reaction from a pathologically increased reaction of BP are not yet available. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²²

In the present study, we calculated the morning surge of BP using Kario’s formula. The early morning surge of BP was not significantly higher in the group of patients with ST depression as compared with patients without these ECG changes. An increased morning surge phenomenon was therefore not associated with transient myocardial ischemia in our patients. A rise of systolic BP especially in the elderly has been described several times as the trigger of ST depression. ⁵ , ⁶ , ²³ , ²⁴ , ²⁵ A possible explanation for this is that the early morning surge is a calculated measure comprising a time interval of several hours. However, ST depression only lasts for a few minutes in most cases. As our results show, hemodynamic parameters are suitable to characterize ST depression only when they are recorded directly during the period of the ST depression.

In general, an increased association with end organ damage is discussed for nondippers. The extent to which the prevalence of ST depression in hypertensive patients is influenced by the day‐night depression of systolic BP has not yet been adequately investigated. Kurpesa and colleagues ²⁶ observed a tendency to more ST depressions in nondippers than in dippers in coronary patients with and without arterial hypertension (n=115). Pierdomenico and colleagues ²⁷ reported that the frequency of ST depression in hypertensive patients with coronary heart disease is only affected at night by the dipping status. In the present study, the prevalence of ST depression was highest (25%) in risers. However, the difference was not significant compared with dippers, extreme dippers, risers, and nondippers. Significantly greater age and significantly higher BPs may have been responsible for the tendency to higher prevalence of ST depression in risers. Both the age and the severity of arterial hypertension are factors that influence the prevalence of ST depression. In the present study, an unequivocal correlation between the prevalence of ST depression and the dipping status could not be shown. However, in a previous work we demonstrated in a larger study population the highest prevalence for ST depression in extreme dippers (28.6%), followed by risers, nondippers, and dippers (21.8%, 19.6%, and 18.2%, respectively). ¹⁴

Several study groups reported a rise of BPs as a trigger of ST depression. ⁵ , ⁶ , ²⁸ We have therefore attempted to establish whether there are fundamental differences in BP parameters between dippers (+extreme dippers) and nondippers (+risers) with intermittent ST depression before, during, and after ST depression (Figure 2). The systolic 24‐hour ABPM was significantly higher in nondippers (+risers) with intermittent ST depression than in dippers (+extreme dippers) with intermittent ST depression (140±19 mm Hg vs 130±12 mm Hg; P≤.05). There was a significant rise of the pressure‐frequency product even before the ST depression in the 2 groups as compared with the respective 24‐hour mean value. During ST depression, all parameters (BP systolic, BP diastolic, heart rate, and pressure‐frequency product) rose significantly in the group of dippers (+extreme dippers) and heart rate and double product in the group of nondippers (+risers). After the ST depression, these parameters had fallen again to ranges comparable to those of the corresponding 24‐hour mean values in both groups. These results confirm corresponding findings of other investigators who postulate that ST depressions are triggered by peaks of BP. No differences in principle were detected in our study in the triggering of ST depression between dippers (+extreme dippers) and nondippers (+risers). This result is surprising in that dippers (+extreme dippers) often have ST‐depression episodes in the evening. However, these are not triggered by dipping of BP, but by rises of BP.

With the same investigative procedure, ST‐depression episodes in the early hours of the morning (6 am to 9 am) were compared with those of the early evening (6 pm–9 pm) independent of the dipping status in order to see whether the time of day affects the hemodynamic triggering mechanisms. ST depression in the period from 6 am to 9 am were accompanied by significantly higher BPs than ST depression between 6 pm and 9 pm, before, during, and after the ST depression. There was a significantly higher pressure‐frequency product and significantly higher diastolic BP during the ST depression. These results suggest that depending on the time of day, the intermittent ischemias detected by means of ST‐segment analysis possibly have different ischemia thresholds. This suspicion was also expressed in the past by other study groups who, for methodologic reasons, were only able to investigate the heart rate at the time of ST depression because they did not have ABPM data. ²⁹ , ³⁰ , ³¹

Conclusions

To what extent these findings must be taken into consideration in planning chronotherapy must be established in further studies with greater sample size and statistical power.

Acknowledgment:  The authors acknowledge Miriam Brieger in the preparation of this manuscript.