This cross-sectional, observational, single-center cohort study included 647 consecutive patients who simultaneously underwent 24-hour ambulatory BP monitor (ABPM) and echocardiography between January 2010 and March 2015. We included patients with normal renal function who were between 18–80 years of age. Patients with systemic inflammatory diseases, significant hepatic or renal dysfunction, neurologic disorders, secondary hypertension, valvular heart disease, atrial fibrillation, a history of heart failure, a history of acute coronary syndrome, or systemic or local malignancies were not included in the study. Demographic characteristics recorded at the first visit included age, sex, height, weight, current medications, smoking history, and other current diseases. Body mass index (BMI) was calculated as the ratio of dry weight in kilograms to height squared (in meters). This study was approved by the Institutional Review Board. All patients were required to provide written informed consent to participate.

Complete blood cell counts including total white blood cells (neutrophils and lymphocytes) and high sensitivity C-reactive protein (hs-CRP) were obtained at the first visit. Total serum cholesterol, triglycerides, high-density lipoprotein and low-density lipoprotein cholesterol, blood glucose, creatinine, and uric acid were also measured. The NLR was calculated as the ratio of the neutrophil count to the lymphocyte count.

Office BP measurements were measured twice at five-min intervals using an automated office BP monitor (HBP-1100-E, OMRON Healthcare, Hoofddorp, the Netherlands). Noninvasive 24-h ABPM was performed on each patient’s non-dominant arm using an automatic oscillometric device (TONOPORT V, PAR Medizintechnik, Berlin, Germany) on a normal working day. Patients were generally asked to refrain from intense exercise and to stop taking antihypertensive medications 24 hours prior to the office visit. All subjects were instructed to rest or sleep between 10:00 PM and 7:00 AM (nighttime) and to continue their usual activities between 7:00 AM and 10:00 PM (daytime). The accuracy of the device was checked against the standard auscultatory method to ensure that the difference in BP measurements between methods did not exceed 5 mm Hg. The device was set to obtain BP readings at 20-min intervals during the daytime and 40-min intervals during the nighttime. Only 24-hour recordings that included at least 80% successful recordings were accepted as valid. Each ABPM dataset was first automatically scanned to remove artifactual readings based on preselected editing criteria. The following ABPM parameters were evaluated: 24-hour mean systolic and diastolic BP levels, daytime mean systolic and diastolic BP levels, nighttime mean systolic and diastolic BP levels, and BP variability assessed by standard deviation (SD). Additionally, the magnitude of the nocturnal decline in BP (Δ nocturnal decline) was calculated as follows: daytime average BP - nighttime average; the percentage change in BP from day to night (% day - night BP) was calculated as: (daytime BP - nighttime BP) × 100 / daytime BP.

Following the recommendations of the European Society of Hypertension,11) a normotensive state was defined as a mean daytime ambulatory systolic and diastolic BP < 135/85 mm Hg and a mean nighttime ambulatory systolic and diastolic BP < 120/70 mm Hg by ABPM. True hypertension was diagnosed if the average daytime BP was higher than 135/85 mm Hg and the average nighttime BP was above 120/70 mm Hg. Because we divided groups by the ABPM parameter, masked hypertensive patients were assigned to the hypertensive group and white coat hypertension patients were assigned to the normotensive group. Hypertensive subjects who showed a < 10% reduction in BP from daytime to nighttime were defined as “non-dippers,” and those who had a reduction in BP ≥ 10% from daytime to nighttime were considered “dippers.” Patients were classified according to the ABPM.

Standard 2-dimensional echocardiography was performed using a 3.5-MHz transducer (Philips iE33, Philips Medical Systems, Bothell, WA, USA) on all subjects while they were lying in the left lateral decubitus position, and examiners were blinded to patient information. Measurements of the thickness of the interventricular septum and posterior wall, the diameter of the left ventricle (LV) cavity, and the LV mass index (LVMI) were performed according to criteria outlined by the American Society of Echocardiography.12) Left atrial volume was measured by prolate ellipse methods. Echocardiographic assessments of EFT, defined as the echo-free space between the outer wall of the myocardium and the visceral layer of the pericardium, were measured perpendicularly from the free wall of the right ventricle at end-systole in three cardiac cycles as previously described.4) Because one of the critical issues in EFT measurement is the inconsistency in measurement location, the mean EFT was averaged from images of the parasternal long axis, parasternal short axis, and apical 4-chamber view. Offline measurement of EFT was performed by two cardiologists (KI Cho and BJ Kim) who were blinded to the clinical data. As previously described, the intra- and inter-observer variabilities of EFT were 3.3% and 5.8%, respectively.

Statistical analyses were performed with the commercially available computer program SPSS 18.0 for Windows (SPSS Inc., Chicago, IL, USA). Data are presented as mean ± SD for continuous variables and percentage (%) for categorical data. The Mann-Whitney U test was used for continuous variables and the chi-square test was used for categorical data. The normality of data was tested using the Kolmogorov-Smirnov test. Parameter differences among the 3 groups were evaluated using a one-way ANOVA for normally distributed variables or the Kruskal-Wallis test for non-normally distributed variables. Relationships between variables were examined with Pearson correlation coefficients. The cutoff values of EFT and NLR for predicting non-dippers with corresponding sensitivity and specificity were estimated by receiver operator characteristic (ROC) curve analysis. Multivariate logistic regression models for non-dippers were designed to determine the variables independently associated with this status. A two-tailed p-value < 0.05 was considered statistically significant.

A total of 535 hypertensive patients and 112 normotensive patients were analyzed, and their clinical features and ambulatory BP parameters according to diurnal variation are shown in Tables 1 and 2. NLR was the highest in the non-dippers compared to the other two groups (non-dipper, 2.75 ± 2.81; dipper, 2.01 ± 1.32; control, 1.92 ± 1.11; p < 0.001) (Table 1, Fig. 1A). The non-dipper group included more females and more patients on beta-blockers (all p < 0.05). Circadian BP profiles and BP variability assessed by 24-hour mean BP SD were greater in hypertensive patients, especially in dippers (all p < 0.05) (Table 2).

Although there was no significant difference in systolic function, hypertensive patients showed significantly greater wall thickness, greater LVMI, and a larger left atrial diameter, all of which were more prominent among non-dippers (Table 3). The mean EFT was significantly higher in both hypertensive groups compared to the control group and was highest in the non-dipper group (non-dipper group, 7.3 ± 3.0 mm; dipper group, 6.1 ± 2.0 mm; control group, 5.6 ± 2.0 mm; p < 0.001) (Table 3, Fig. 1B).

EFT was significantly correlated with age (r = 0.160, p < 0.001) (Fig. 2A), BMI (r = 0.091, p = 0.042) (Fig. 2B), 24-hour mean BP variability (r = 0.152, p = 0.001) (Fig. 2C), and NLR (r= 0.353, p < 0.001) (Fig. 2D). NLR was also significantly correlated with 24-hour mean BP variability (r = 0.270, p = 0.001) (Fig. 2E).

Furthermore, an EFT ≥ 7.0 mm was associated with the non-dipper BP pattern with 51.3% sensitivity and 71.6% specificity [ROC area under curve of 0.606, 95% confidence interval (CI) 0.56–0.65, p < 0.001] (Fig. 3A). And NLR ≥ 2.1 was also associated with non-dipper BP pattern with 52.2% sensitivity and 65.3% specificity (ROC area under curve of 0.596, 95% CI 0.55–0.64, p < 0.001) (Fig. 3B).

In multivariate analysis, EFT [adjusted odds ratio (OR) = 3.985, 95% CI = 1.215–13.066, p = 0.022] and NLR (OR = 1.341, 95% CI = 1.052–1.710, p = 0.018) were independent parameters that distinguished the non-dipper pattern after adjustment for CV risk factors (Table 4).