This cross-sectional study included patients from two centers (West China Hospital and Chengdu Seventh People's Hospital) in China. Adult patients were eligible if they (1) had CKD, (2) were diagnosed with hypertension, and (3) agreed to undergo 24 h ABPM. Exclusion criteria were patients (1) on dialysis (hemodialysis or peritoneal dialysis), (2) with history of malignancy, (3) with <70% valid records on 24 h ABPM, and (4) who were pregnant. Patients with albuminuria (albumin/creatinine ratio ≥30 mg/g) or estimated glomerular filtration rate (eGFR) <60 ml/min/1.73 m² or abnormalities of kidney structure for over 3 months were diagnosed with CKD (17). eGFR was determined from serum creatinine levels using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (18). The study protocol was approved by the ethics committee of the West China Hospital, Sichuan University, and was approved by the Institutional Review Board. Written informed consent was obtained from all patients.

A total of 447 CKD patients formed the cohort of this cross-sectional study. Baseline evaluation including medical history, physical examination, and laboratory tests were recorded at the time patients visited the outpatient clinic or were admitted to hospital. These data included sex, age, body mass index (BMI), current alcohol consumption, current smoking, CV history, antihypertensive drugs, diabetes mellitus (DM), biochemical parameters, urinary protein test, and two-dimensional echocardiogram.

Experienced nurses measured office BP when participants were admitted to our hospital or visited the outpatient clinic. After patients were allowed to rest quietly for 5–10 min, the mean value of three consecutive BP measurements at 5-min intervals using a mercury sphygmomanometer was recorded. The 24-h ABPM was performed using Space labs 90217 devices (Space labs Medical, Redmond, WA, USA) at West China Hospital and an ABPM 6100 (Welch Allyn, Chicago, IL, USA) at Chengdu Seventh People's Hospital, with BP readings set at 20-min intervals from 6:00 a.m. to 10:00 p.m., and 30- or 60-min intervals from 10:00 p.m. to 6:00 a.m. Day- and nighttime BPs were defined as mean BP during the period from 6:00 a.m. to 10:00 p.m. and 10:00 p.m. to 6:00 a.m., respectively. Patients were instructed to undergo their usual activities and take antihypertensive drugs as usual and were encouraged to sleep no later than 10:00 p.m. and to get up at approximately 6:00 a.m. ABPM was not conducted if the patient worked nightshifts. The time patients went to bed and woke up was recorded. A measurement with at least 70% of diurnal and nocturnal BP readings was regarded as a successful ABP. The ABP must be taken within 3 days after the measurement of office BP. Both office BP and ABP measurements were taken from the non-dominant arm with an appropriate cuff size based on arm circumference at the time of enrollment.

Echocardiography was performed by two experienced ultrasonologists according to standardized procedures, as previously reported (19). The linear method was applied to quantify LV mass (20): LVM (g) = 0.8 × 1.04 × [(interventricular septum + LV internal diameter + posterior wall thickness)³ – LV internal diameter³] + 0.6 g. Left ventricular hypertrophy (LVH) was defined according to the 2015 American Society of Echocardiography/European Association of Cardiovascular Imaging chamber quantification document, with LVMI per body surface area >95g/m² in women and 115 g/m² in men (19).

Morning hypertension was defined as mean BP ≥135/85 within 2 h after waking (11). Morning surge in blood pressure was defined as SBP during the 2-h period immediately after waking minus the average of three SBP readings centered around the lowest nighttime SBP value (21). Patients were divided into the elevated morning surge group (≥15 mmHg) and low morning surge group (<15 mmHg) according to the third quantile. Dipping patterns of BP and heart rate (HR) were calculated according to the following formula: mean night/day ratio of SBP, DBP, and HR. Patients were diagnosed as dippers if the ratio was >0.8–0.9, non-dippers if the ratio was >0.9–1, or reverse dippers if the ratio was >1 (22). Patients were defined to have achieved the goal for ambulatory BP when 24-h, daytime, and nighttime BP was <130/80, <135/85, and <120/70 mmHg, respectively, and to have the goal for office BP when the BP was <140/90 (23). CV history was defined as history of heart disease, peripheral vascular disease, or cerebral vascular disease.

Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp, Armonk, NY, USA) and GraphPad Prism 8.0.1. Normally distributed data are presented as mean ± standard deviation, skewed data as median values with interquartile range, and categorical variables as numbers with percentage. Log-transformation or square root calculations were applied for normal transformation. Data were analyzed using the chi-square test or Fisher's exact test for categorical variables, the Student's t-test for normally distributed data, and the Wilcoxon rank-sum test for continuous skewed variables.

Receiver-operating characteristic (ROC) curve analysis was conducted, and the area under the curve (AUROC) was calculated to assess the ability of night blood pressure and morning blood pressure surge to predict morning hypertension. Binary logistic regression was applied to detect the factors associated with morning hypertension and to calculate the risk of LVH correlated with morning hypertension. Linear regression analysis was used to identify the association between morning hypertension and eGFR, UPCR, and LVMI. The regression β-coefficient represented the contribution of the independent variables to the dependent variables. To determine whether the correlation between morning hypertension and TOD was independent of night hypertension or morning surge, we added night hypertension and morning surge to the regression model in models 2 and 3, respectively. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated, and a two-tailed p < 0.05 was considered statistically significant.

In total, 583 patients were enrolled. A total of 136 were excluded because of the following factors: complication with malignant tumors (5), incomplete ABPM data (50), and dialysis (81). Finally, 447 patients with CKD and hypertension were included in this study. Two hundred (44.7%) patients were male, and the mean age was 67 ± 14 years. A total of 194 (43.4%) participants had morning hypertension, among which 186 (95.9%) had nocturnal hypertension, and 97 (50%) had morning surge in blood pressure with ≥15 mmHg. A total of 320 (71.6%) participants had nocturnal hypertension, and the morning surge in 152 (34%) patients was ≥15 mmHg.

Baseline features of the participants grouped according to morning BP are shown in Table 1. No differences in age, sex, current smoking, current alcohol consumption, CV history, or DM was found between the two groups. Compared with those with morning normotension, the proportion of use of calcium channel blockers, α blockers, and over two kinds of antihypertensive drugs was higher in patients with morning hypertension. Higher levels of white blood cells, neutrophils, creatinine, uric acid, and phosphorus were observed in the morning hypertension group, whereas the levels of lymphocytes, total protein, albumin, and calcium were lower in this group (Table 1). Regarding the BP parameters, control of 24 h, daytime, and nighttime BP and that measured at the clinic were all poorer in the morning hypertension group (p < 0.001). No differences were found in SBP, DBP, or HR dipping between the two groups (Table 2).

Association between nighttime blood pressure and morning surge in blood pressure with morning hypertension.

Among patients in the morning hypertension group, only 4.1% had normal nighttime BP, compared with 47% in the morning normotension group (p < 0.001). Moreover, 97 (50%) patients with morning hypertension had elevated morning surge, while the rate was 21.7% in patients with morning normotension (p < 0.001) (Table 2). According to the multivariate logistic regression model, nighttime SBP (OR, 1.2; 95%CI, 1.11–1.3) and morning surge (OR, 1.2; 95%CI, 1.15–1.26) were independently correlated with morning hypertension (p < 0.001) (Table 3).

To assess the ability to use nighttime BP and morning surge in blood pressure for predicting morning hypertension, ROC curve analysis was performed (Figure 1). Nighttime SBP could better distinguish morning hypertension (AUROC, 0.85; 95%CI, 0.81–0.88; p < 0.001), compared with nighttime DBP (AUROC, 0.75; 95%CI, 0.71–0.88; p < 0.001) and morning surge (AUROC, 0.69; 95%CI, 0.65–0.74; p < 0.001). The optimal cutoff points for morning hypertension were 130 mmHg for nighttime SBP (sensitivity, 73.2%; specificity, 83%) and 70 mmHg for nighttime DBP (sensitivity, 62%; specificity, 75%).

According to univariate analysis, both morning hypertension and nocturnal hypertension were shown to be associated with higher level of LVMI and UPCR, lower eGFR (p < 0.01), and LVH (p < 0.001), whereas only higher level of UPCR was found to correlate with elevated morning surge in blood pressure (p = 0.026) (Supplementary Tables 1–3).

According to multivariate linear and logistic regression analysis, morning hypertension correlated strongly with lower level of eGFR (β = −0.51; 95%CI, −0.95–−0.08; p < 0.05), higher level of LVMI (β = 0.06; 95%CI, 0.04–0.08; p < 0.001), UPCR (β = 0.22; 95%CI, 0.06–0.38; p < 0.01), and LVH (OR, 2.14; 95%CI, 1.3–3.51; p < 0.01) when nocturnal hypertension and elevated morning surge in blood pressure were considered (Table 4, model 3). After further adjustments of baseline features (age, sex, BMI, current alcohol consumption, current smoking, DM, and CV history), biochemical indices, and office BP control, the significance did not change. When morning hypertension was considered, TOD was not shown to be associated with nocturnal hypertension and elevated morning surge in blood pressure, except for UPCR, which was higher in patients with nocturnal hypertension (Table 4). However, the difference was not significant after adjustment of biochemical indices and office BP.

To further assess the correlation between morning BP and TOD, morning SBP and DBP were analyzed as continuous variables in the regression model. Both morning SBP and DBP were found to be associated with LVH, higher UPCR, and LVMI (Figure 2, Supplemental Table 4). Regarding eGFR, only morning SBP was shown to be negatively correlated with eGFR (all p < 0.01), and the correlation between morning DBP and eGFR was only significant in model 4 (Figure 2, Supplementary Table 4).