Introduction

Uncontrolled hypertension causes target organ damage (TOD) in the heart, brain, eyes, arteries and kidneys, and the coexistence of TOD and hypertension increases the overall cardiovascular risk.1 Chronic kidney disease (CKD), which is either a consequence or cause of hypertension, is frequently observed in patients with hypertension, and substantially accelerates the hypertension-related risk of cardiovascular events.1, 2, 3, 4 In addition, hypertension is the second most common cause of end-stage renal disease, and it aggravates the age-related decline of renal function if blood pressure (BP) is not adequately controlled.5

Current guidelines recommend a strict BP control (<130/80 mm Hg) in patients with CKD, although the target of BP remains a matter of debate.1, 2 Despite the need for strict BP control in patients with CKD, past evidence shows that a mere 27% of CKD patients achieve a BP of <140/90 mm Hg and even fewer (11%) reach a BP of <130/85 mm Hg.6 In Korea, nationwide studies suggested office BP control (<130/80 mm Hg) in the general population and Type 2 diabetes was inadequate (58.5% and 39.9%, respectively).7, 8 We can predict that the BP of a large amount of patients with CKD or at an increased risks for CKD is not adequately controlled in Korea, although there was no direct information regarding the control status of BP in Korean CKD patients.

CKD is associated with not only elevated office BP but also high ambulatory BP (ABP) (for example, masked hypertension) and a lack of diurnal BP variation.9, 10 Several studies reported that high ABP and/or an abnormal circadian BP variation (that is, non-dipper or reverse-dipper) could predict combined cardiovascular disease and renal outcomes better than office BP values in patients with CKD.11, 12, 13, 14, 15 In Korea, Song et al.16 reported that nocturnal declines in the BP do not occur in patients with renal failure and that the impairment of circadian rhythm occurred in the early stage of renal dysfunction even though BP was well-controlled. However, these data were driven by a small number and are not representative of the Korean population.

In this nationwide survey, using office BP and ABP monitoring (ABPM), we examined the BP control status and nocturnal BP dipping pattern in hypertensive patients with CKD. We also examined the clinical factors associated with such BP pattern.

Methods

Study design

The Assessment of Blood Pressure Control and Target Organ Damage in Patients with Chronic Kidney Disease and Hypertension (APrODiTe) study was a nationwide, cross-sectional study conducted at 21 centers between October 2009 and May 2011. The primary objective was to identify the distribution of BP control categories, such as ‘true controlled,’ ‘white-coat,’ ‘masked’ and ‘sustained uncontrolled hypertension,’ and dipping statuses, such as ‘extreme-dipper,’ ‘dipper,’ ‘non-dipper’ and ‘reverse-dipper.’ Secondary objectives were to evaluate (1) TOD, according to BP control status (estimated glomerular filtration rate (eGFR), proteinuria (spot urine protein-to-creatinine ratio >300 mg g−1 Cr or alb ⩾1+ at urinalysis, left ventricular hypertrophy (LVH) and heart failure)) and (2) factors related to each BP control pattern.

The study included patients who (1) provided informed consent, (2) were aged 20–75 years, (3) were diagnosed with hypertension for >6 months (or⩾3 clinic visits) prior to participation and had taken antihypertensive drugs for >3 months, (4) had an eGFR (calculated using the MDRD study equation) between 15 and 89 ml min−1 per 1.73 m2 and (5) had good compliance to medication and no change in prescription 2 weeks prior to participation. Diabetic nephropathy was defined according to the clinical or pathological diagnosis by clinicians. We clinically diagnosed diabetic nephropathy if patients had a prolonged duration of diabetes, impaired renal function or proteinuria, and if there was no evidence of other diseases. We did not include patients who had an eGFR over 60 and no proteinuria. All participants had abnormalities in renal function, urinalysis, pathology or imaging studies. The study excluded patients who (1) changed prescription according to ABPM regularly, (2) had acute kidney injury/hospitalization, (3) had a proteinuria of >6 g day−1 (spot urine protein-to-creatinine ratio of >6.0 g per g Cr), (4) had end-stage renal disease with dialysis and/or were a kidney recipient, (5) had diseases, such as uncontrolled arrhythmia, uncontrolled bronchial asthma/chronic obstructive pulmonary disease or primary endocrinologic diseases except diabetes mellitus, (6) were pregnant/lactating women or (7) were night-shift workers.

We estimated the sample size using the following equation (95% confidence interval and 2.5% confidence interval length).

The estimate is the largest under a prevalence of 0.5. Therefore, the sample size for this study was calculated to be 1537.

The protocol was approved by the institutional review boards of the participating centers. Data collected included clinic BP, ABP, anthropometric measures, prescribed drugs, past medical history (including the incidence of coronary artery disease and heart failure) and laboratory data including serum creatinine, eGFR and morning urine protein-to-creatinine ratio. LVH was defined using the Romhilt–Estes point score system (‘definite’ ⩾5 points; ‘probable’ 4 points), because only 31 participants (2.35%) performed echocardiography in this study. Previous studies reported that the Romhilt–Estes scoring system generated the most favorable predictive values for LVH.17, 18

Clinic BP measurement

All clinic BP measurements were performed by trained staff using an oscillometric OMRON IA-2 automatic BP device (IntelliSense, Omron, Kyoto, Japan). Three consecutive seated BP readings were recorded at intervals of 1–2 min, and the clinic BP reading was the mean of the last two readings.

ABP measurement

Twenty-four-hour ABPM was performed using an oscillometric TM-2430 monitor (A&D, Seoul, Korea). The monitor was programmed to record BP every 30 min. The ABP readings were considered adequate if the monitor had been worn for 24 h and if there were ⩾16 acceptable readings between 0800 hours and 2200 hours (daytime) and ⩾12 acceptable readings between 2200 hours and 0800 hours (nighttime). We used the ABP definitions proposed by Fagard et al.19

Definitions

Controlled clinical BP was defined as a level <140/90 mm Hg. ABP was considered normal if the daytime value was <135/85 mm Hg and the nighttime value was <120/70 mm Hg.

True controlled hypertension was defined as a controlled clinic and ambulatory (daytime and nighttime) BP. Masked hypertension was defined as controlled clinic BP and elevated daytime/nighttime ABP. Sustained uncontrolled BP was defined as uncontrolled clinic and ABP. The nocturnal dipping pattern was defined as a ratio of the mean nighttime systolic BP (SBP) to the mean daytime SBP. The patients were classified as extreme-dippers, if the ratio was less than 0.8; dippers, if the ratio was between 0.8 and 0.9; non-dippers, if the ratio was between 0.9 and 1.0; and reverse-dippers, if the ratio was greater than 1.0.

Statistical analysis

The baseline characteristics were compared between the patients of different BP categories using an ANOVA/Kruskal–Wallis test or the χ2-test, as appropriate. Multiple logistic regression models were used to relate the BP category status to demographic factors, anthropometric data and laboratory findings. Continuous variables were expressed as the mean±s.d. (parametric)/median (non-parametric), and categorical variables were expressed as n (%). All P-values were two-sided and were considered significant if P<0.05.

Results

A total of 1545 patients were enrolled and 1317 patients were evaluated (Figure 1).

Figure 1
figure 1

Diagram of participant enrollment and analysis. A total of 1545 patients were enrolled, and 228 patients were excluded (n=1,317). Of the 228 patients excluded from the analysis, 102 (44.7%) patients could not complete the ambulatory BP monitoring.

Full size image

Baseline characteristics

The median age of the analyzed patients was 58 years and 62.9% of the patients were men. Diabetic nephropathy was reported in 298 (25.0%) patients. Hypertensive nephropathy, chronic glomerulonephritis and polycystic kidney disease were reported in 510 (38.7%), 289 (21.9%) and 27 (2.1%) patients, respectively. A total of 345 (26.2%) patients were given a single antihypertensive drug, whereas 461 (35.0%) and 294 (22.3%) patients received two and three antihypertensive drugs, respectively. As many as 1171 (88.9%) patients took angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (ARBs). Beta-blockers were used in 33.9% of all patients, calcium channel blockers in 56.3% and diuretics in 37.5%. The median serum creatinine was 1.45 mg dl−1, and the median eGFR was 48 ml min−1 per 1.73 m2 (CKD stages 2, 3 and 4: 416 (31.6%), 640 (48.6%) and 261 (19.8%), respectively).

Hypertension control categories

Among participants, the mean clinic BP was 137.4/81.6 mm Hg. As assessed via ABPM, the mean daytime BP was 135.7/80.9 mm Hg, and the mean nighttime BP was 122.9/73.4 mm Hg. According to the definitions of clinic and ABP, 703 (53.4%) patients had controlled clinic BP, whereas 504 (38.3%) patients had controlled ambulatory daytime or nighttime BP. Sustained uncontrolled hypertension was encountered most frequently (n=557, 42.3%), followed by masked hypertension (n=447, 33.9%), true controlled hypertension (n=256, 19.4%) and white-coat hypertension (n=57, 4.3%). Twenty-four-hour SBP was the highest in patients with sustained uncontrolled hypertension, followed by those with masked hypertension, white-coat hypertension and true controlled hypertension (Table 1 and Figure 2).

Table 1 Characteristics by hypertension control categories and dipping patterns
Full size table
Figure 2
figure 2

Comparison of clinic and ambulatory BP in each BP control and dipping category. (a) The 24-hour SBP was the highest among patients with sustained uncontrolled hypertension followed by those with masked, white-coat and true controlled hypertension. (b) Although there were no significant differences in clinic SBP and daytime SBP (except among extreme-dippers) among the four groups, nighttime SBP was highest in reverse-dippers (136.4 mm Hg). The nighttime SBP of non-dippers, dippers and extreme-dippers followed that of the reverse-dippers (130.2, 127.4 and 126.9 mm Hg, respectively). *P<0.05.

Full size image

The prevalence of dippers and non-/reverse-dippers on the hypertension control categories were as follows: true controlled, 119 (46.5%) vs. 99 (38.7%); white-coat, 29 (50.9%) vs. 8 (14.1%); masked, 120 (26.8%) vs. 267 (59.7%); and sustained uncontrolled, 171 (30.7%) vs. 308 (55.3%) (P<0.001). When we compared the night SBP/day SBP ratio according to the hypertension control categories, the values in the true controlled, white-coat, masked and sustained uncontrolled hypertension were 0.88±0.078, 0.82±0.067, 0.92±0.111 and 0.92±0.111, respectively (P<0.001). In conclusion, masked and sustained uncontrolled hypertension was associated with a loss of dipping (non-/reverse-dipper).

The percentage of men was the highest among patients with sustained uncontrolled hypertension and the lowest among those with true controlled hypertension. The highest number of current smokers was among patients with sustained uncontrolled hypertension.

Diabetic nephropathy and an LVH score ⩾4 were the most prevalent factors in patients with sustained uncontrolled hypertension. A medical history of stable angina, acute coronary syndrome and heart failure was similar among the four hypertension control categories. The eGFR was highest among patients with true controlled hypertension (51.85 ml min−1 per 1.73 m2 vs. 48.20, 45.80 and 43.00 ml min−1 per 1.73 m2 in masked, sustained uncontrolled and white-coat hypertension, respectively), and the magnitude of proteinuria was the highest among patients with sustained uncontrolled hypertension (620.0 mg g−1 Cr vs. 376.0, 230.9 and 180.0 mg g−1 Cr in masked, true controlled and white-coat hypertension, respectively). Patients with sustained uncontrolled hypertension were prescribed more antihypertensive drugs than other patients.

True controlled hypertension was the most frequently encountered pattern among patients with CKD stage 2 (23.1%, P=0.03), whereas sustained uncontrolled hypertension was most frequently encountered among those with CKD stage 4 (48.3%, P=0.05) (Table 2).

Table 2 Hypertension control patterns according to renal function
Full size table

Dipping patterns

The prevalence of non-dippers was the highest (n=454, 34.5%), followed by dippers (n=439, 33.3%), reverse-dippers (n=228, 17.3%) and extreme-dippers (n=196, 14.9%). The prevalence of dippers among patients with CKD stages 2, 3 and 4 were 158 (38.0%), 203 (31.7%) and 78 (29.9%), respectively (P<0.05) (Table 2). Although there were no significant differences in clinic SBP or daytime SBP (except among extreme-dippers) between the four groups, nighttime SBP was highest in reverse-dippers (141.4 mm Hg) (Table 1 and Figure 2).

Age, gender and duration of hypertension were not different between the four groups. The prevalence of diabetic nephropathy, stable angina and heart failure were higher in reverse-dippers than in patients of the other groups. The eGFR was highest in extreme-dippers (53.0 ml min−1 per 1.73 m2) and the prevalence of CKD stage 4 was highest in reverse-dippers (n=59, 25.9%). Proteinuria differed according to the dipping status. The proteinuria amount of extreme-dippers, dippers, non-dippers and reverse-dippers were 662±948.4, 744±1001.5, 1096±1362.1 and 1163±1446.3 mg g−1 Cr, respectively (P<0.001). Reverse-dippers were prescribed more antihypertensive drugs than other patients. Non-dippers and reverse-dippers were taking more beta-blockers and calcium channel blockers than dippers and extreme-dippers. The use of diuretics was similar among all groups.

Association between BP burden and TOD

Young age (<58 years), female gender and the presence of non-diabetic CKD were independently related to true controlled hypertension. The use of angiotensin-converting enzyme inhibitor or ARB was also associated with true controlled hypertension, even though this association was not statistically significant (Table 3). Old age (⩾58 years), male gender, high body mass index (BMI) and diabetic CKD were independently related to sustained uncontrolled hypertension. The use of beta-blocker and calcium channel blocker was also significantly associated with sustained uncontrolled hypertension. When we analyzed BP patterns according to renal function, sustained uncontrolled hypertension was found to be significantly related to CKD stage 4, female gender, lower BMI and diabetic CKD (Table 4).

Table 3 Factors related with each BP control and dipping category
Full size table
Table 4 Effect of each BP control and dipping category on renal dysfunction
Full size table

Table 3 also presents data on factors associated with the categories of dipping patterns. A high BMI, non-diabetic CKD and current smoking were independently associated with extreme-dippers. Extreme-dippers showed low levels of proteinuria. Diabetic nephropathy and the use of beta-blockers were significantly associated with being a reverse-dipper. A high BP burden (non-dippers and reverse-dippers), old age, low BMI and diabetic nephropathy were independently associated with CKD stages 3 and 4 (Table 4).

Discussion

This study demonstrated a high prevalence of masked/sustained uncontrolled hypertension and non-/reverse-dippers in Korean patients with CKD. Uncontrolled BP and less nocturnal BP dipping in CKD patients were associated with low eGFR, higher levels of proteinuria and pre-existing heart disease.

Several studies from Japan, China and the USA also reported the associations of BP control status or ABP values with TOD.20, 21, 22 The results of these studies were concordant with those of this study regarding a significantly large proportion of higher BP load (masked, sustained uncontrolled hypertension/non-, reverse-dipper) in CKD patients and a meaningful association between a higher BP load and severe TOD, including renal dysfunction and cardiovascular disease. However, there was a small difference between the studies in the proportion of hypertension categories and dipping status. These differences may arise from the difference in baseline characteristics such as ethnicity, diabetes, distribution of CKD stages and diuretic use.

Ethnicity has an impact on BP control and dipping status. Ethnicity can also be related to genetic variation, environmental factors such as diet, smoking and alcohol, or an interaction between these components. Several reports found various allele frequencies of BP-related genes such as ACE/angiotensinogen genes and their association with responsible proteins.23, 24 These variations can lead to varied effects in ARB/-ACE inhibitor treatment in various ethnicities. Furthermore, polymorphisms were related with hypertension itself and TODs such as LVH. Sacks et al. reported that SBP was associated with salt intake and that the DASH diet had an increased positive effect especially in the high-salt group. Both genetics and environmental factors, therefore, are associated with differences in BP control and dipping status between various ethnicities.25

The prevalence of diabetes was approximately 35, 5.4 and >22.6% in the Japanese, Chinese and our study, respectively. Participants in the Japanese, American and our study used diuretics in 27, 82.4 and 37.5%, respectively. In addition, analyses using variables such as the presence of nocturia and smoking status as measured via a simple questionnaire may affect the results.

In patients with CKD, more strict BP control is recommended to reduce TOD. However, it is difficult to follow this recommendation because of several mechanisms including extracellular fluid volume expansion, the activation of the rennin angiotensin system and increased sympathetic nervous system activity, among others, leading to the loss of BP circadian rhythm.26, 27, 28, 29, 30, 31 Thus, clinic BP provides an incomplete and potentially misleading assessment of the severity of hypertension. An assessment of the actual BP with ABPM may be useful in the pharmacological management of hypertension in patients with CKD to slow the progression of renal impairment and reduce the development of cardiovascular disease.32 As shown in the present study, the use of ABPM reveals a high prevalence of masked hypertension and a high number of non-/reverse-dippers among the Korean population, which is in turn associated with TOD.

Several mechanisms have been proposed for elevated nighttime BP and a high BP burden in patients with CKD. These mechanisms include an expanded extracellular fluid volume caused by a decreased renal excretion of sodium and water with a linked elevation of oxidative stress as well as increased sympathetic nervous system activity.27, 28, 29, 30, 31 The activation of the renin angiotensin system in CKD patients is also a critical factor leading to the loss of dipping.30, 31

The capacity to excrete sodium during the day is a significant determinant of nocturnal BP and dipping.22 Non-dipping is a compensatory phenomenon to produce pressure natriuresis in volume-overloaded patients. Diuretic treatment has been shown to have a beneficial role in modulating the circadian rhythm of BP.29, 33, 34 Furthermore, dietary potassium supplementation and sodium restriction can restore normal dipping and enhance the effect of nighttime BP drug therapy.35 Only 37.5% of patients used diuretics as an antihypertensive drug in this study. The pattern of antihypertensive drug choice can be related to a higher prevalence of patients with masked and sustained uncontrolled hypertension and increased numbers of non-/reverse-dippers in this study.

Previous studies have suggested that the alteration of the timing of antihypertensive drug therapy may reduce nighttime BP.36 A simple modification such as switching the dosing of an ARB from morning to evening may be all that is needed.37 Several studies have reported a decrease in the night/day ratio as a result of changing the time of dosing of one of the antihypertensive agents from the morning to the evening.31, 38 Studies have also shown that urinary albumin excretion is also significantly reduced after bedtime treatment with ARB39 and that the progression of nephropathy was delayed.40 Given that non-dipping is a potential independent risk factor for CKD progression and the development of cardio-renal syndrome, optimizing antihypertensive therapy (pharmacological or dietary) may be relevant. Unfortunately, we could not deduce an association between the timing of antihypertensive drug therapy and the dipping pattern.

In this study, beta-blockers and calcium channel blockers were used in patients with a high BP burden (sustained uncontrolled hypertension, reverse-dipper). This may be due to the practice pattern of physicians prescribing angiotensin-converting enzyme inhibitors or ARBs first, especially in cases of diabetic nephropathy. However, further studies are required to evaluate the effects of specific classes of antihypertensive drugs on BP burden and dipping.

Home-measured BP is superior to clinic-measured BP in reducing the misclassification of hypertension caused by white-coat and masked hypertension41 as well as in predicting CKD-associated complications.10 Several experts have recommended increasing the use of home BP monitoring for routine care and suggested that this should be supported by health-care systems. In this study, 44.7% of the excluded patients from the analysis failed to properly carry out ABPM. Although ABPM may provide more detailed information, home BP monitoring, which is easier and more accessible, may be better for assessing BP patterns in high-risk patients such as those with CKD.

To the best of our knowledge, this is the largest study of ABPM in patients with CKD, and in which the clinic BP measurements were carefully standardized and performed by a trained staff using frequently calibrated equipment. However, our study has some limitations. First, we obtained data concerning heart problems such as stable angina, acute coronary syndrome and heart failure through the patients’ medical history based on patients’ recall and medical records; therefore, there may be a significant lack of data. Second, we did not fully evaluate the LVH or LV mass index because only a small proportion of patients underwent echocardiography (n=31 (2.3%)). We therefore may have underestimated the relationship between each BP category or dipping pattern and heart problems. Third, ABPM was evaluated once; thus, the variability may result in some degree of misclassification. Fourth, our study did not assess sleep quality, which influences nocturnal BP. We were unable to draw a longitudinal relationship between any BP category and TOD because of the cross-sectional nature of the analyses.

In conclusion, clinic BP potentially mischaracterized patients’ BP burden because of increased nighttime BP. ABPM revealed a significantly high prevalence of masked/sustained uncontrolled hypertension and a large number of non-/reverse-dippers, which may be associated with hypertension-related TOD. Korean physicians should attempt to acquire more data regarding patients’ BP through frequent home BP or ABPM. Korean physicians should also be more careful in drug choice and chronotherapy to reverse their patients’ dipping pattern to normal. Future studies are warranted to assess the predictive values of ABP for cardiorenal events in Korean CKD patients.