Relationship Between Morning Hypertension Identified by Home Blood Pressure Monitoring and Brain Natriuretic Peptide and Estimated Glomerular Filtration Rate: The Japan Morning Surge 1 (JMS-1) Study

Authors


Kazuomi Kario, MD, PhD, COE (Center Of Excellence) Program, Division of Cardiovascular Medicine, Department of Medicine, Jichi Medical University, 3311-1, Yakushiji, Shimotsuke, Tochigi 329-0498, Japan
E-mail: kkario@jichi.ac.jp

Abstract

We evaluated whether morning minus evening systolic blood pressure (SBP) difference (MEdif) in home blood pressure measurements can be a marker for hypertensive target organ damage. The authors analyzed 611 hypertensive patients who had high morning SBP levels (≥135 mm Hg). The patients with morning hypertension (MEdif ≥15 mm Hg, average of morning and evening SBP [MEave] ≥135 mm Hg) were older (P<.001) and had a longer duration of hypertension and antihypertensive medication use, a higher prevalence of left ventricular hypertrophy (LVH) on electrocardiography, a lower glomerular filtration rate by the Cockcroft-Gault equation (P=.002), and a higher brain natriuretic peptide (BNP) level (P<.001) than those with well-controlled blood pressure (MEdif <15 mm Hg, MEave <135 mm Hg). The patients with morning hypertension had a higher BNP level than those with well-controlled blood pressure after adjustment for the confounding factors (28.7 pg/mL vs 20.0 pg/mL; P=.033). In conclusion, morning hypertension is more likely seen among patients with older age and longer duration of hypertension and antihypertensive medication use, and it may be associated with a higher prevalence of LVH and a higher BNP level.

Cardiovascular events tend to occur frequently in the morning.1 Elevation of morning blood pressure (BP) levels has been reported to increase the risk for cardiovascular mortality in the general population.2 We reported that a morning systolic BP (SBP) surge on ambulatory BP monitoring was a risk factor for cerebrovascular disease independent of 24-hour BP level.3 It is difficult to perform ambulatory BP monitoring in all patients,4 although ambulatory BP, which includes BP during physical activity and sleeping time, is reported to be a somewhat better cardiovascular prognostic measure than is home BP.5 Therefore, as a substitute for ambulatory BP, we evaluated whether the difference of morning minus evening SBP (MEdif) in home BP measurements is related to hypertensive target organ damage in medicated hypertensive patients on a stable drug regimen with elevated morning SBP levels (>135 mm Hg).

We previously reported that in the Jichi Morning Hypertension Research (J-MORE) study,6 morning SBP measured by self-measured BP monitoring at home remained uncontrolled in 60.4% of 969 stably medicated hypertensive outpatients. The highest quartile of MEdif was >15 mm Hg, and the determinants were age, regular alcohol drinkers, and use of β-blockers.6 We did not measure any markers of hypertensive target organ damage in the J-MORE study, and it remained unclear whether exaggerated MEdif (>15 mm Hg) is associated with hypertensive target organ damage.

The Japan Morning Surge 1 (JMS-1) study was a randomized controlled study aiming to clarify whether strict control by doxazosin of morning BP measured using home BP monitoring can decrease brain natriuretic peptide (BNP) levels and urinary albumin excretion ratio adjusted by creatinine (UAR) level. The detailed protocol of the JMS-1 study is described elsewhere.7 BNP is not only a marker of left ventricular hypertrophy (LVH) and cardiac overload; Wang and colleagues8 showed that a small increase in plasma BNP level was a risk factor for future stroke events, transient ischemic attacks, and cardiovascular mortality in a general population without a history of heart failure. As a substudy of baseline data in the JMS-1 study, we evaluated here whether MEdif is related to UAR and BNP. These will be useful preliminary data for evaluating whether reduction of the MEdif using an α-blocker or another antihypertensive agent taken just before going to bed can decrease UAR and BNP, although a previous study9 with a long-acting nondihydropyridine calcium channel blocker (CCB) taken at night to reduce morning BP did not show a difference in cardiovascular outcome compared with a group that did not receive this agent.

METHODS

Inclusion Criterion of the JMS-1 Study

In the baseline data for the JMS-1 study, we selected hypertensive outpatients who had uncontrolled morning SBP (>135 mm Hg). The patients had maintained a stable antihypertensive medication status for at least 3 months. Morning BP was measured via self-measured BP monitoring at home according to the 2004 Japanese Society of Hypertension (JSH) Guidelines for Management of Hypertension.10 BP values were recorded by the patients and reported to their own physicians. The average morning SBP level of the first measurements for 3 days (3 points) within 1 week before visiting the physician's office was used as the value for evaluating whether patients met the criterion for enrollment.

Exclusion Criteria

We excluded patients who had arrhythmias, history of heart failure with or without hospital admission, presence of orthostatic hypotension, dementia, and presence of malignancy or chronic inflammatory disease or who were taking an α-blocker or a β-blocker. The histories of the patients were collected by the physicians who were managing the patients in their clinic or hospital. The diagnosis of heart failure was based upon clinical history and results of physical examination at enrollment.

Patients' Backgrounds

The definition of a habitual drinker was a patient who reported that he or she drank alcohol more than 5 days per week regardless of the amount of alcohol. Smoking was defined as having a current smoking habit. Diabetes mellitus was defined as treatment for diabetes mellitus or a fasting blood glucose level >7.0 mmol/L (126 mg/dL) or a casual glucose level >11.1 mmol/L (200 mg/dL) in patients who were not treated. Glucose intolerance was defined as a fasting blood glucose level from 6.1 to 6.9 mmol/L (110–125 mg/dL). Hyperlipidemia was defined as a total cholesterol level >5.7 mmol/L (220 mg/dL) or a triglyceride level >1.7 mmol/L (150 mg/dL). Clinical histories of the patients were obtained from interviews by the patients' own doctors. Antihypertensive medications were classified as CCBs, angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), diuretics, and others.

Guidelines for Self-Measurement of BP at Home

All the self-measured home BP readings were measured using a validated cuff oscillometric device, the HEM-705IT (Omron, Kyoto, Japan).11 BP measurements at home were performed according to the 2004 JSH guidelines.10 Patients were instructed to place the cuff on the same arm for all measurements and to measure BP in a sitting position after more than 2 minutes of rest. BP measurements were made daily in a seated position twice per occasion with a 15-second interval for 3 days (6 points). Patients were instructed to write down all of the measured values of BP and pulse rates and to report them to their physicians. Morning BP was measured within 1 hour after waking, after urination, before breakfast, and before taking antihypertensive medication. Evening BP was measured before going to bed, and the patients were instructed to avoid measuring BP just after taking a bath, drinking alcohol, or smoking.

Classification of Patients According to Home SBP Level

According to the home BP levels, we classified the patients into 4 groups: well-controlled group (average of morning and evening SBP [MEave] <135 mm Hg and MEdif <15 mm Hg), morning BP predominant group (MEave <135 mm Hg and MEdif ≥15 mm Hg), sustained hypertension group (MEave ≥135 mm Hg and MEdif <15 mm Hg), and morning hypertension group (MEave ≥135 mm Hg and MEdif ≥15 mm Hg). There was no available evidence about the significance of different levels of MEdif; therefore, we used 15 mm Hg, because the highest quartile of MEdif in the J-MORE study was >15 mm Hg.6

Clinic BP

Clinic BP was measured in the office in a similar way as self-measured BP at home. Physicians or nurses measured clinic BP in the office.

Blood and Urinary Examinations

Blood samples were drawn from a vein in the morning in a fasting state. Blood examinations were performed at the time of enrollment. We measured the level of plasma BNP with radioimmunoassay (Shionogi Research Laboratories, Osaka, Japan). The covariance of measurements of BNP was 5.20% to 6.11%. The estimated glomerular filtration rate (eGFR) was calculated afterward using the Modification of Diet in Renal Disease (MDRD) study equation modified for Japanese patients12: eGFR-MDRD (mL/min/1.73 m2) [0.741 × 175 × Age−0.203× Scr−1.154× (0.742 if female)]. Chronic kidney disease was defined as eGFR-MDRD <60 mL/min/1.73m2. The eGFR was also calculated afterward using the Cockcroft-Gault equation (eGFR-CG). Spot samples of urine were collected in the morning. We assessed the existence of proteinuria and UAR at the time of enrollment in the study. Urinary microalbumin levels were measured by an immunoturbidimetric method (Mitsubishi Kagaku Iatron, Tokyo, Japan). Covariance of measurements of microalbumin was 0.66% to 8.75%.

Electrocardiography

Electrocardiography was performed using validated machines of the participating institutes. Physicians reported the presence of LVH as assessed by Sokolow-Lyon criteria (SV1+RV5>35 mm and strain pattern).

Statistical Analysis

We enrolled 617 patients at the beginning of the present study. After excluding 6 patients (morning SBP <135 mm Hg at the time of enrollment: 3 patients; α-blocker or β-blocker users, 2 patients; duplication of enrollment: 1 patient), statistical analyses were conducted for 611 patients. Data are shown as mean ± SD. BP analysis was conducted using the average of 2 measurements per day for 3 days (6 points) obtained in a seated position. Fifty-four patients had morning SBP (the average of the 6 points) <135 mm Hg because the SBP level of these patients was remarkably reduced at the second measurement on each occasion. MEdif was calculated as the difference of morning SBP minus evening SBP. MEave was calculated as the average of morning SBP and evening SBP. The values of BNP and UAR skewed deviations. The analyses of BNP and UAR values were performed for the log-transformed value and analyzed as continuous variables. Correlations among log10BNP, log10UAR, eGFR, and BP levels were analyzed using Pearson's correlation. MEdif and MEave were thereby found to be related to some variables; therefore, analysis of covariance was performed and partial correlation coefficients were determined to rule out their effects on these parameters in multiple linear regression analysis. One-way analysis of variance was performed to detect differences among groups, and Tukey's honestly significant differences test was used for multiple pairwise comparisons of means among groups. Analysis of covariance was performed to detect the groups with differences among BP groups after adjustment for confounding factors, and the Bonferroni test was used for multiple pairwise comparisons. The computer software SPSS version 11.0 (SPSS Inc, Chicago, IL) was used for the analyses and probability <.05 was considered statistically significant.

RESULTS

Patients' Background and BP Levels

The average age of the patients for whom baseline data were collected in the JMS-1 study was 70.2±9.6 years (range, 30–92 years); 340 were male and 271 were female. The average body mass index and waist-to-hip ratio were 24.1±3.4 kg/m2 and 0.91±0.09, respectively. Current smokers comprised 116 (19.0%) patients and habitual alcohol drinkers comprised 196 (32.1%) patients. The percentages of the patients with past illness were 11.6% for cerebrovascular events and 8.2%for ischemic heart disease. The average number of antihypertensive drugs that had been used by the patients was 1.63±0.74 (CCBs, 66.1%; ACEIs, 14.2%; ARBs, 58.4%; diuretics, 21.8%). The average BP levels (systolic/diastolic) were 152±14/82±10 mm Hg in the morning, 140±16/76±10 mm Hg in the evening, and 156±18/83±11 mm Hg in the clinic, respectively. MEdif ranged from −29 to 78 mm Hg (mean, 11±13 mm Hg) and MEave was 146±13 mm Hg. The backgrounds of the patients with MEdif <0 mm Hg, MEdif 0 to 15 mm Hg, and MEdif ≥15 mm Hg are shown in Table I.

Table I.  Backgrounds of the Patients With MEdif <0 mm Hg, MEdif 0–15 mm Hg, and MEdif >15 mm Hg
 MEdif <0 mm Hg (n=105)MEdif 0–15 mm Hg (n=284)MEdif >15 mm Hg (n=222)P Value
Age, y67.3±11.969.8±9.472.1±8.1a,b<.001
Male, %42.945.443.7.88
Body mass index, kg/m224.8±3.324.2±3.623.7±3.1c.016
Waist-to-hip ratio0.91±0.070.91±0.090.91±0.09.72
Current smoking, %21.018.019.4.79
Habitual drinking, %26.729.238.3.041
Duration of hypertension, y10.8±8.611.2±9.212.0±9.4.45
Duration of antihypertensive medication, y7.1±7.18.1±7.29.1±8.5.083
No. of antihypertensive drugs1.7±0.71.6±0.71.7±0.7.040
Any CCB use, %67.668.262.6.39
Long-acting CCB use, %60.059.054.3.49
Short- or intermediate-acting CCB use, %9.58.88.7.97
ACEI use, %15.213.415.1.84
ARB use, %59.056.462.1.44
Diuretic use, %25.715.927.9.003
Nighttime administration of antihypertensive drugs, %15.212.012.3.69
Nitrates, %1.93.54.1.59
Diabetes or impaired glucose tolerance, %16.216.514.9.87
Hyperlipidemia, %40.035.936.0.74
Fasting blood glucose, mg/dL97.8±22.099.7±23.299.2±22.7.78
Hemoglobin A1c, %5.2±0.65.3±0.85.2±0.7.31
Serum creatinine, mg/dL0.81±0.310.80±0.280.78±0.20.73
eGFR-CG, mL/min75.2±35.270.0±25.965.3±20.9b.005
eGFR-MDRD, mL/min/1.73 m266.5±20.065.3±14.864.7±14.6.63
Geometric mean UAR, mg/g creatinine28.132.126.8.29
Geometric mean BNP, pg/mL22.022.728.8a,c.009
Clinic SBP, mm Hg155±18154±18159±19a.022
Clinic DBP, mm Hg85±1283±1283±11.26
Clinic pulse rate, beats per minute71±1272±1270±11.080
MEave SBP, mm Hg150±14d145±13145±13e.002
MEave DBP, mm Hg83±1179±977±8<.001
MEave pulse rate, mm Hg69±1069±868±8.41
Sokolow-Lyon LVH, %8.67.912.4.24
aP<.05 vs patients with MEdif 0–15 mm Hg. bP<.001; cP<.05 vs patients with MEdif <0 mm Hg. dP<.01 vs patients with MEdif 0–15 mm Hg. eP<.01 vs patients with MEdif <0 mm Hg. Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BNP, brain natriuretic peptide; CCB, calcium channel blocker; DBP, diastolic blood pressure; eGFR-GC, estimated glomerular filtration rate by Cockcroft-Gault equation; eGFR-MDRD, estimated glomerular filtration rate by Modification of Diet in Renal Disease equation; LVH, left ventricular hypertrophy; MEave, average of morning and evening measurements; MEdif, difference of morning minus evening measurements; SBP, systolic blood pressure; UAR, urinary albumin excretion rate/creatine. Data are shown as mean ± SD or percentage. UAR and BNP are shown as geometric mean. Overall P values were calculated by analysis of variance. Intergroup diffences were calculated by Tukey's honestly significant differences test.

Correlation of BP Levels and BNP, eGFR, and UAR Values

Log10BNP was significantly correlated with MEdif (r=0.13; P=.001) and MEave (r=0.18; P<.001). In multiple linear regression analysis, the correlation between log10BNP and MEdif remained significant (β=0.16; P<.001) after adjustment for MEave. In addition, log10BNP tended to correlated with MEdif (β=0.07; P=.063) even after adjustment for MEave, age, body mass index, duration of antihypertensive medication, number of antihypertensive drugs, diuretic use, and eGFR-CG in multiple linear regression analysis. On the other hand, log10UAR did not correlate with MEdif (r=–0.01; P=.78), while it correlated with MEave (r=0.22; P<.001).

There was a significant correlation between eGFR-CG and MEdif (r=−0.13; P=.001) and MEave (r=−0.18; P<.001). The correlation between eGFR-CG and MEdif remained significant even after adjustment for MEave in multiple linear regression analysis (β=−0.15; P<.001); however, the correlation between eGFR-CG and MEdif disappeared after adjustment for age and body mass index (P=.22). There was no significant correlation between eGFR-MDRD and MEdif.

Morning Hypertension

The prevalence of morning hypertension (MEave ≥135 mm Hg and MEdif ≥15 mm Hg), sustained hypertension (MEave ≥135 mm Hg and MEdif <15 mm Hg), and morning BP predominant in the present study are shown in Figure 1. The characteristics of the patients with well-controlled BP, morning BP predominant, sustained hypertension, and morning hypertension are shown in Table II.

Figure 1.

Prevalence of morning hypertension in the Japan Morning Surge 1 (JMS-1) study. Well-controlled blood pressure (BP) group (average of morning and evening measurements [MEave] <135 mm Hg and morning systolic BP (SBP) minus evening SBP values [MEdif] <15 mm Hg), morning BP predominant group (MEave <135 mm Hg and MEdif ≥15 mm Hg), sustained hypertension group (MEave ≥135 mm Hg and MEdif <15 mm Hg), and morning hypertension group (MEave ≥135 mm Hg and MEdif ≥15 mm Hg).

Table II.  Characteristics of Patients With Morning Hypertension
 Well-Controlled (n=68)Morning Predominant (n=51)Sustainedhypertension (n=321)Morning Hypertension (n=171)P Value
MEdif (median), mm Hg<15 (8.7)>15 (24.3)<15 (3.5)>15 (22.5) 
MEave (median), mm Hg<135 (131.1)<135 (128.8)>135 (146.4)>135 (148.0) 
Age, y66.2±11.970.3±8.269.7±9.7a72.6±7.9a,b,c<.001
Male, %39.754.945.840.4.23
Body mass index, kg/m224.1±3.823.3±3.124.4±3.523.8±3.2.08
Waist-to-hip ratio0.89±0.080.91±0.090.91±0.080.91±0.09.23
Current smoking, %14.719.619.619.3.82
Habitual drinking, %26.543.129.036.8.07
Duration of hypertension, y10.1±7.69.4±8.011.3±9.312.8±9.6.048
Duration of antihypertensive medication, y7.1±6.56.8±6.88.0±7.39.8±8.8.016
No. of antihypertensive drugs1.5±0.71.7±0.81.6±0.71.7±0.7.13
Any CCB use, %66.266.068.461.5.50
Long-acting CCB use, %55.962.060.052.1.34
Short- or intermediate-acting CCB use, %7.46.09.49.5.83
ACEI use, %13.214.014.115.4.97
ARB use, %50.054.058.664.5.18
Diuretic use, %16.228.019.127.8.06
Nighttime administration of antihypertensive drugs, %10.36.013.414.2.41
Nitrates, %4.44.02.84.1.83
Diabetes or impaired glucose tolerance, %16.29.816.516.4.67
Hyperlipidemia, %38.225.536.839.2.35
Fasting blood glucose, mg/dL99.2±22.896.7±18.299.2±22.9100.0±23.9.85
Hemoglobin A1c, %5.3±0.85.1±0.55.3±0.85.3±0.7.34
Serum creatinine, mg/dL0.76±0.200.79±0.230.81±0.300.78±0.19.44
Geometric mean UAR, mg/g creatinine28.9 (8.5–98.1)20.7 (5.6–77.0)31.5 (8.2–120.3)28.9 (7.6–110.5).22
eGFR-MDRD, mL/min/1.73 m267.8±15.267.0±14.565.1±16.664.1±14.5.35
Chronic kidney disease, %33.829.434.839.2.57
Clinic SBP, mm Hg147±16148±13156±18a,e162±19b,d,f<.001
Clinic DBP, mm Hg82±1181±1184±1283±11.35
Clinic pulse rate, beats/min71±1169±1272±1270±11.20
Sokolow-Lyon LVH, %5.93.98.514.7.036
aP<.05; bP<.001 vs well-controlled group. cP<.05; dP<.01 vs sustained hypertension group. eP<.01, fP<.001 vs morning predominant group. Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CCB, calcium channel blocker; DBP, diastolic blood pressure; eGFR-MDRD, estimated glomerular filtration rate by Modification of Diet in Renal Disease equation; LVH, left ventricular hypertrophy; MEave, average of morning and evening measurements; MEdif, difference of morning minus evening measurements; SBP, systolic blood pressure; UAR, urinary albumin excretion rate/creatine. Data are shown as mean ± SD or percentage. Values of UAR are shown as geometric mean (SD range). P values were calculated by analysis of variance and Tukey's honestly significant differences or chi-squared test.

Differences in plasma BNP level among the BP groups are shown in Figure 2. Even after adjustment for significant confounding factors such as age, duration of hypertension and antihypertensive medication, habitual alcohol drinking, diuretics use, and eGFR-CG, the patients with morning hypertension had significantly higher plasma BNP levels than those with well-controlled BP (28.7 pg/mL vs 20.0 pg/mL; P=.033).

Figure 2.

Brain natriuretic peptide (BNP) level in the patients with morning hypertension. Well-controlled blood pressure (BP) group (average of morning and evening measurements [MEave] <135 mm Hg and morning systolic BP (SBP) minus evening SBP values [MEdif] <15 mm Hg), morning BP predominant group (MEave <135 mm Hg and MEdif ≥15 mm Hg), sustained hypertension group (MEave ≥135 mm Hg and MEdif <15 mm Hg), and morning hypertension group (MEave ≥135 mm Hg and MEdif ≥15 mm Hg). Plasma BNP levels are shown as geometric mean. Overall differences were calculated by analysis of variance, and intergroup differences were calculated by Tukey's honestly significant differences test. Significance was indicated by P<.05.

There were no significant differences in UAR level among the patients in the 4 BP groups. Differences in eGFR-CG among the 4 BP groups are also shown in Figure 3. The difference of eGFR among the 4 BP groups disappeared after adjustment for age and body mass index (P=.45), however. There were no statistical differences in eGFR-MDRD among the 4 BP groups.

Figure 3.

Estimated glomerular filtration rate (eGFR) calculated using the Cockcroft-Gault equation in patients with morning hypertension. Overall difference was calculated by analysis of variance, and intergroup differences were calculated by Tukey's honestly significant differences test. Significance was indicated by P<.05. BP indicates blood pressure.

DISCUSSION

We have demonstrated that morning hypertension detected by home BP monitoring was related to older age, longer duration of hypertension and antihypertensive medication use, increased BNP level, higher prevalence of electrocardiographically identified LVH, and decreased eGFR-CG level. The relationship between morning hypertension and increased BNP level remained significant even after adjustment for the significant confounding factors. These results indicate that morning hypertension is more likely seen among patients with older age and longer duration of hypertension and antihypertensive medication use and may be associated with higher prevalence of LVH and a higher BNP level.

The patients with morning hypertension had a higher BNP level than those with well-controlled BP after adjustments for confounding factors. Morning BP level in home BP is reported to be a better prognostic parameter for cardiovascular mortality than office BP,2 and Ikeda and colleagues13 demonstrated that left ventricular mass index was increased with an increase in MEdif of home BP. The relationship observed here between morning hypertension and BNP level may be related to the mechanisms of increased risk of morning hypertension.

We focused on MEdif in the present study, but the patients with morning hypertension had significantly higher clinic BP levels than those in the other 3 groups. Clinic BP might have contributed to higher BNP levels and electrocardiographically identified LVH among the patients with morning hypertension. It has been reported that ambulatory BP predicts cardiovascular events after adjustment for clinic BP in untreated14 and treated hypertension,15 and Bobrie and associates16 reported that masked hypertension detected by home BP monitoring was a good prognostic marker of cardiovascular mortality. The home BP level is more predictive of cardiovascular risk than is clinic BP level. Therefore, in the present study, we classified the patients using only home BP assessment. Further studies among patients with the same clinic BP levels may be needed to clarify the relationship between morning hypertension and hypertensive target organ damage.

The risk of patients with higher evening BP levels than morning BP levels was not clarified in the present study. Evening BP values were reported to be higher than morning BP values in a study of untreated hypertension in the Netherlands.17 Asayama and coworkers18 reported that morning BP and evening BP measurements provided equally useful information for stroke risk in the Japanese general population. In the present study, the patients with MEdif <0 mm Hg (the patients whose evening SBP level was higher than the morning SBP level) had a lower BNP level than those with MEdif 0 to 15 mm Hg and those with MEdif >15 mm Hg, although the patients with MEdif <0 mm Hg had a higher MEave level and a similar clinic SBP level compared with those with MEdif 0 to 15 mm Hg. The patients with MEdif <0 mm Hg were younger and heavier in the present study. It has been reported that obese patients had lower BNP levels,19 and this might affect the results of the present study. Obesity is an important risk factor for cardiovascular events, and decreased BNP levels may not always be associated with lower cardiovascular risk in obese patients.

The patients with morning predominant BP did not have a significant difference in BNP level compared with the patients with well-controlled BP. The patients with morning predominant BP were older and had a higher percentage of diuretic use, and a higher percentage were habitual drinkers. MEdif per se may be a BP variation influenced by age, and the risk of MEdif may have additive effects in patients who have an increased MEave level.

Plasma BNP level has been reported to be affected by age20 and renal function.21 In the present study, eGFR-CG level was significantly lower in the patients with morning hypertension; however, these relationships might be derived from the differences in age and body mass index and have little to do with morning BP. Therefore, eGFR-MDRD did not have significant differences in the 4 BP groups. MEdif or morning hypertension was not related to UAR, a marker of hypertensive renal damage, in the present study. UAR was correlated to the same degree with morning, evening, and office BP level. Kamoi and associates22 reported, however, that morning BP level in home BP was related to UAR in diabetic patients. We enrolled patients with elevated morning BP levels; they were at high risk for hypertensive renal damage. BP overload to the kidney throughout the day might have been more important than BP change during the day.

Study Limitations

Elevated morning BP level includes 2 different types of BP variation: nocturnal hypertension type (such as nondippers) and morning BP surge type.23 We were unable to distinguish between these 2 BP types, using conventional home BP devices, among the patients with exaggerated MEdif. We excluded patients who had a history of heart failure at enrollment; we did not perform echocardiography in all patients.

CONCLUSIONS

Morning hypertension is more likely seen among patients with older age and longer duration of hypertension and antihypertensive medication use, and it may be associated with higher prevalence of LVH and a higher BNP level.

Acknowledgments:

Participants and participating centers included Shizukiyo Ishikawa, Jichi Medical University of School of Medicine; Kazuo Eguchi, Jichi Medical University of School of Medicine, Shioya General Hospital and Sano Municipal Hospital; Toru Hashimoto, Jichi Medical University of School of Medicine; Masato Morinari, Jichi Medical University of School of Medicine; Satoshi Hoshide, Jichi Medical University of School of Medicine and Oyama Municipal Hospital; Yoko Hoshide, Sato Clinic; Motoyuki Ishiguro, Ishiguro Clinic; Toshio Nakayama, Nakayama Clinic; Hideo Hirose, Kosakacho Clinic; Naoshi Yamada, Yamada Brain Surgery Clinic; Akira Yoshimura and Makoto Yamashita, Mishima Clinic; Masanori Harada and Hitoshi Nishimura, Nishiki Central Hospital; Ruri Kaneda, Jyosai Hospital; Yoshio Matsui and Seiichi Shibasaki, Miwa Municipal Hospital; Mitsunobu Murata, Koga Red Cross Hospital; Joji Ishikawa, Koga Red Cross Hospital and Sano Municipal Hospital; and Yasuyuki Mizumori, Ieshima Clinic and Uzuka Clinic (20 physicians and 16 institutes).

Ancillary