The association between plasma homocysteine and ambulatory blood pressure variability in patients with untreated hypertension

Clinica Chimica Acta 477 (2018) 32–38

Contents lists available at ScienceDirect

Clinica Chimica Acta journal homepage: www.elsevier.com/locate/cca

The association between plasma homocysteine and ambulatory blood pressure variability in patients with untreated hypertension

T

Suhua Lia,1, Jieming Zhua,b,1, Lin Wub,1, Long Penga, Yanting Luoa, Yunyue Zhaoa, Ruimin Donga, ⁎ ⁎⁎ Lin Chena, Xixiang Tangc, , Jinlai Liua, a

Department of Cardiology, the Third Affiliated Hospital, Sun Yat-sen University, Tian-he Road, Guangzhou 510630, China Department of Electrocardiography, the Third Affiliated Hospital, Sun Yat-sen University, Tian-he Road, Guangzhou 510630, China c Advanced Medical Center, the Third Affiliated Hospital, Sun Yat-sen University, Tian-he Road, Guangzhou 510630, China b

A R T I C L E I N F O

A B S T R A C T

Keywords: Homocysteine Hyperhomocysteinemia Blood pressure variability Hypertension Stroke

Background: Both homocysteine (Hcy) and blood pressure variability (BPV) are independent predictors of stroke, however, their relationship is rarely evaluated before. This study aimed to investigate the association Hcy and ambulatory BPV in subjects with untreated primary hypertension. Methods: A total of 252 eligible patients were recruited. Plasma Hcy was measured and 24-h ambulatory blood pressure monitoring was performed for each subject. The systolic and diastolic BPV values were calculated as the SD of individual blood pressure values during 24 h, daytime and nighttime, and then stratified by the tertiles of Hcy concentration (T1 to T3). Univariate and multivariate linear regression models were used to assess the relationships between Hcy tertiles and BPV variables. Results: The mean values of Hcy from T1 to T3 were 7.51 ± 1.21 μmol/l, 11.09 ± 1.07 μmol/l and 19.14 ± 6.26 μmol/l, respectively. Systolic and diastolic mean blood pressures were similar among subjects with different Hcy tertiles. However, both systolic and diastolic BPV variables, no matter in 24-h, daytime or nighttime, were increasing significantly along with the rises in Hcy tertiles (all p < 0.05 for linear trends analysis). Multivariate linear regression analysis indicated that Hcy tertiles were significantly associated with BPV variables, independently of mean blood pressures other confounding factors. In subgroup analysis, the associations between Hcy tertiles and BPV variables were enhanced by the increased risk stratification of hypertension. Conclusions: Plasma Hcy was positively and independently associated with ambulatory BPV in patients with untreated hypertension.

1. Introduction Homocysteine (Hcy) is a non-proteinogenic α-amino acid synthesized from methionine, involving in a consequence of the biochemical reactions [1]. An abnormally elevated concentration of plasma Hcy, conventionally > 15 μmol/l, is defined as hyperhomocysteinemia (HHcy), which can lead to endothelial cell injury, vascular inflammation and atherogenesis [2]. Therefore, HHcy is considered to be an independent risk factor for atherosclerotic and thromboembolic diseases [1–3]. Previous studies suggested that HHcy could significantly increase the risk of stroke, especially in hypertensive patients [3,4]. Reduction of HHcy with folic acid therapy is proved to be highly effective in primary prevention of stroke among adults with hypertension [5].

On the other hand, current studies have demonstrated that blood pressure variability (BPV) is also an independent predictor of primary stroke in hypertensive patients [6,7]. BPV can induce vascular dysfunction, arterial stiffness and atherosclerosis, which shares the proposed pathophysiologic mechanisms of HHcy [7,8]. Therefore, it is very interesting to verify the relationship between Hcy and BPV in patients with hypertension. Unfortunately, the establishment of their association is rarely investigated before.

⁎

Correspondence to: X. Tang, Advanced Medical Center, the Third Affiliated Hospital, Sun Yat-sen University, Tian-he Road, Guangzhou 510630, China. Corresponding author. E-mail addresses: [email protected] (X. Tang), [email protected] (J. Liu). 1 These authors contributed to the work equally. ⁎⁎

https://doi.org/10.1016/j.cca.2017.11.042 Received 11 June 2017; Received in revised form 19 November 2017; Accepted 30 November 2017 0009-8981/ © 2017 Published by Elsevier B.V.

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

2. Materials and methods

performed to assess the non-adjusted relationships between plasma Hcy tertiles and BPV, with the first tertile as the reference. Then, two multivariate linear regression models were performed to adjust the confounding factors. The first model was adjusted for age, gender, BMI, history of diabetes mellitus, smoking status, total cholesterol, uric acid, creatinine, HbA1c, and hemoglobin. The second model was additionally adjusted for the corresponding mean blood pressure values in order to investigate whether the observed associations were independent of the mean blood pressure. In order to model linear trends, we also ran models in which Hcy tertiles were entered as ordinal variables. Separate models were established for both SBPV and DBPV during 24 h, daytime, and nighttime, respectively. Subgroup analysis was performed based on the cardiovascular risk stratification of hypertension recommended by 2013 ESC/ESC guidelines [9]. A 2-tailed P < 0.05 was considered to indicate statistical significance.

2.1. Study population The present study was a prospective study approved by the Institutional Review Board of the hospital. Consecutive inpatients with untreated essential hypertension, who presented to the Department of Cardiology between January 2016 and March 2017 for blood pressure monitoring and cardiovascular risk stratification, were recruited. Subjects were eligible to participate in the study if they: (1) were ≥ 18 y; (2) were diagnosed as primary hypertension, with systolic blood pressure (SBP) ≥ 140 mmHg and/or diastolic blood pressure (DBP) ≥ 90 mmHg at rest based on the mean of two or more readings on three separate office visits; (3) had no history of using antihypertensive medications; (4) did not received any testing of Hcy before recruiting. Those with secondary hypertension, acute or chronic infectious diseases, heart failure, acute cardiocerebrovascular events, thyroid disorder, anemia, mental illness or malignant tumor were excluded. The written informed consents for included subjects were obtained.

3. Results 3.1. Patient characteristics A total of 252 consecutive patients (including 136 males and 116 females) with untreated essential hypertension were recruited. Overall, the mean age of subjects was 62.6 ± 12.7 y old, with mean BMI of 24.91 ± 3.29 kg ∙ m− 2. The values of Hcy ranged from 4.53 to 39.99 μmol/l for all subjects, with a mean of 11.58 ± 6.13 μmol/l. Among them, 69 (27.4%) had HHcy (> 15 μmol/l). Baseline characteristics were stratified by Hcy tertiles in Table 1. The values of Hcy ranged from 4.53 to 9.45 μmol/l in T1 subjects, 9.56 to 12.85 μmol/l in T2 subjects, and 13.06 to 39.99 umol/l in T3 subjects. The mean values of Hcy from T1 to T3 subjects were 7.51 ± 1.21, 11.09 ± 1.07, and 19.14 ± 6.26 μmol/l, respectively (p < 0.001). More males (p < 0.001) and smokers (p < 0.001) were seen in T2 and T3 subjects when compared with T1 subjects. Percentage of diabetes was higher in T3 subjects than those in T1 and T2 subjects (p = 0.039). Increasing trends form T1 to T3 subjects were observed in the concentrations of serum creatinine (p = 0.007), uric acid (p = 0.008), and hemoglobin (p = 0.024). However, there were no significant differences in mean age, body mass index, lipid profiles, HbA1c, staging of hypertension, and cardiovascular risk stratification of hypertension among subjects with different Hcy tertiles.

2.2. Data collection and blood sample measurements The baseline information of included subjects were collected, including age, gender, body mass index, smoking status, and history of diabetes mellitus. Within 24 h after admission, the fasting venous blood samples were collected into evacuated tubes without additives (for preparation of serum), or containing EDTA as anticoagulants (for preparation of plasma). Blood samples were kept cold (< 4 °C) after collection and centrifuged within 1 h after sampling. The serum and plasma supernatant were transferred to new vials and stored at −20 °C until analyzed. Plasma Hcy was measured on the Siemens ADVIA Centaur. Serum concentrations of total cholesterol, triglyceride, highdensity lipoprotein cholesterol, low-density lipoprotein cholesterol, uric acid, creatinine and cystatin C were measured using the Hitachi 7180 chemistry analyzer. Hemoglobin was measured by the cyanmethemoglobin method. HbA1c analyses were performed by HPLC. 2.3. Evaluation of ambulatory BPV On the same day when blood samples were obtained, patients were arranged to receive a 24-h ambulatory blood pressure monitoring using a validated noninvasive device (TM-2430, A&D). The device was programmed to measure blood pressure every 30 min during the 24-h period, for a maximum of 48 measurements. The number of available blood pressure measurements should be ≥80% of expected values. In other words, all subjects in this study had complete data on at least 38 (80%) of the 48 possible measurements. Otherwise, ambulatory blood pressure monitoring would be repeated. Daytime measurements were defined as the measurements that occurred from 6 am to 10 pm, while nighttime measurements were defined as the ones occurred in the remaining period. The mean SBP and DBP were calculated for 24-h, daytime and nighttime periods, respectively. The systolic and diastolic BPV variables (SBPV and DBPV) were calculated as the SD of individual blood pressure values during each period.

3.2. Blood pressure measurements stratified by Hcy tertiles The mean SBP values during 24 h, daytime and nighttime were 141.0 ± 13, 142.1 ± 13.7, and 138.1 ± 15.2 mmHg for all subjects, while the corresponding DBP were 85.3 ± 10.9, 86.6 ± 10., and 83.0 ± 10.2 mmHg. Analysis of variance showed that there were no significant differences in both SBP and DBP among subjects with different Hcy tertiles. (Fig. 1). Meanwhile, the 24-h, daytime and nighttime systolic BPV were 13.2 ± 3.7, 13.9 ± 3.4 and 12.1 ± 4.3 mmHg for all subjects, with the corresponding diastolic BPV of 10.1 ± 3.7, 10.9 ± 3.2, and 9.1 ± 3.2 mmHg. Unlike SBP and DBP, analysis of variance showed that the 24-h and daytime SBPV and DBPV increased significantly along with the rises in Hcy tertiles. For nighttime SBPV and DBPV, no differences were observed between T1 and T2 subjects, but significant increases were seen in T3 subjects when compared with the other two tertiles. Univariate linear trends analysis indicated that Hcy tertiles showed positive and statistically significant linear trends with all of the 24-h, daytime, and nighttime BPV variables. Multivariate analysis further confirmed that the linear trends between Hcy tertiles and BPV variables were independently of the corresponding mean blood pressures and other confounding factors (Fig. 2).

2.4. Statistical analysis Statistical analysis was performed by using SPSS 22.0 for Windows. Categorical variables were expressed as numbers and percentages, while continuous variables were presented as means ± SD. The participants were grouped by the tertiles of Hcy. Baseline characteristics and BPV values were stratified by Hcy tertiles. Differences of continuous variables were evaluated by the student's t-test or analysis of variance, as appropriate. Differences of categorical variables were evaluated by Pearson χ2 test. Univariate linear regression was 33

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

Table 1 Baseline characteristics stratified by Hcy tertiles. Characteristics

Total (n = 252)

T1 (n = 84)

T2 (n = 84)

T3 (n = 84)

p-Value

Hcy, μmol/l Age, y Male, n(%) Body mass index, kg∙ m− 2 Diabetes, n(%) Smoker, n(%) Serum creatinine, μmol/l Uric acid, umol/l Hemoglobin, g/l Total cholesterol, mg/dl Triglyceride, mg/dl HDL-C, mg/dl LDL-C, mg/dl HbA1c, % Risk stratification of hypertension, n(%) Low-to-moderate High Very high

11.58 ± 6.13 62.6 ± 12.7 136 (54.0) 24.91 ± 3.29 75 (29.8) 81 (32.1) 78.9 ± 33.5 395.7 ± 114.7 137.3 ± 14.5 4.71 ± 1.44 1.85 ± 1.25 1.14 ± 0.31 2.88 ± 0.99 6.0 ± 1.2

7.51 ± 1.21 60.9 ± 12.3 18 (21.4) 24.90 ± 3.21 24 (28.6) 9 (10.7) 64.4 ± 13.0 246.6 ± 110.7 131.4 ± 11.4 4.83 ± 1.35 168 ± 1.01 1.19 ± 0.32 2.94 ± 1.12 6.2 ± 1.2

11.09 ± 1.07 62.8 ± 12.7 63 (75.0) 25.51 ± 3.22 18 (21.4) 39 (46.4) 80.3 ± 16.9 400.9 ± 103.8 139.6 ± 14.1 4.74 ± 1.61 2.06 ± 1.81 1.09 ± 0.24 2.83 ± 0.79 6.1 ± 1.5

19.14 ± 6.26 64.29 ± 13.3 55/29 (65.5) 24.14 ± 3.32 33 (39.3) 33 (39.3) 92.0 ± 51.0 439.7 ± 113.7 141.0 ± 16.1 4.57 ± 1.39 1.81 ± 0.66 1.15 ± 0.36 2.88 ± 1.05 5.9 ± 0.7

< 0.001 NS < 0.001 NS 0.039 < 0.001 0.007 0.008 0.024 NS NS NS NS NS NS

90 (35.7) 45 (17.9) 117 (46.4)

24 (9.5) 18 (7.1) 42 (16.7)

36 (14.3) 12 (4.8) 36 (14.2)

30 (11.9) 15 (6.0) 39 (15.5)

Hcy = homocysteine.

DBPV. For nighttime BPV, however, positive relationships were only observed in T3 subjects (β = 0.445, p = 0.003 for nighttime SBPV; and β = 0.388, p = 0.010 for nighttime DBPV), but not in T2 subjects (β = 0.125, p = NS for nighttime DBPV; and β = 0.001, p = NS for nighttime SBPV).

3.3. Univariate and multivariate linear regression analysis Both univariate and multivariate linear regression analysis indicated that statistically significant and positive relationships were observed between most of BPV variables and Hcy tertiles. (Table 2) In model 1, using T1 subjects as the reference group, the standardized βestimates for 24-h SBPV were 0.240 (p = 0.046) in T2 subjects and 0.573 in T3 subjets (p < 0.001), respectively. Additional adjustment for mean blood pressure seemed to enhance these associations. Similar findings were also observed for 24-h DBPV, daytime SBPV and daytime

3.4. Subgroup analysis Subgroup analysis was also performed based on the risk stratification of hypertension. (Fig. 3, Appendix A) In multivariate analysis, Hcy

Fig. 1. Ambulatory blood presure values stratified by Hcy tertiles. (A.B) 24-h mean systolic and diastolic blood pressures (SBP and DBP); (C.D) daytime mean SBP and DBP; (E.F) nighttime mean SBP and DBP. There were no significant differences in both SBP and DBP among subjects with different Hcy tertiles (all p < 0.05). T1 = the first tertile; T2 = the second tertile; T3 = the third tertile.

34

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

Fig. 2. Ambulatory blood presure variability (BPV) values stratified by Hcy tertiles. (A.B) 24-h mean systolic and diastolic BPV (SBPV and DBPV); (C.D) daytime mean SBPV and DBPV; (E.F) nighttime mean SBPV and DBPV. Both systolic and diastolic BPV variables were increasing significantly along with the rises in Hcy tertiles. *p < 0.05 and **p < 0.01 vs. T1 subjects. #p < 0.05 and ##p < 0.01 vs. T2 subjects. Model 1 was adjusted for age, gender, body mass index, history of diabetes mellitus, smoking status, total cholesterol, uric acid, creatinine, HbA1c, and hemoglobin. Model 2 was additionally adjusted for the corresponding blood presure values. T1 = the first tertile; T2 = the second tertile; T3 = the third tertile.

Table 2 Univariate and multivariate linear regression analysis assessing the relationships of Hcy tertiles with BPV. BPV measurements

24-h systolic BPV Univariate Model 1a Model 2b 24-h diastolic BPV Univariate Model 1 Model 2 Daytime systolic BPV Univariate Model 1 Model 2 Daytime diastolic BPV Univariate Model 1 Model 2 Nighttime systolic BPV Univariate Model 1 Model 2 Nighttime diastolic BPV Univariate Model 1 Model 2

T1

T2

T3

β (95%CI)

P-value

β (95%CI)

p-Value

Ref. Ref. Ref.

0.218 (0.003;0.433) 0.240 (− 0.016;0.496) 0.276 (0.033;0.518)

0.047 0.046 0.027

0.614 (0.399;0.829) 0.573 (0.309;0.838) 0.598 (0.348;0.848)

< 0.001 < 0.001 < 0.001

Ref. Ref. Ref.

0.266 (0.058;0.475) 0.238 (0.001;0.476) 0.308 (0.080;0.535)

0.013 0.049 0.009

0.661 (0.452;0.869) 0.692 (0.447;0.937) 0.713 (0.482;0.944)

< 0.001 < 0.001 < 0.001

Ref. Ref. Ref.

0.279 (0.060;0.499) 0.348 (0.084;0.612) 0.379 (0.120–0.638)

0.013 0.010 0.005

0.591 (0.372;0.810) 0.639 (0.367;0.912) 0.655 (0.390–0.921)

< 0.001 < 0.001 < 0.001

Ref. Ref. Ref.

0.316 (0.100;0.532) 0.321 (0.079;0.564) 0.369 (0.119;0.618)

0.005 0.010 0.004

0.613 (0.397;0.830) 0.681 (0.431;0.931) 0.695 (0.446;0.945)

< 0.001 < 0.001 < 0.001

Ref. Ref. Ref.

0.076 (− 0.156;0.307) 0.125 (− 0.151;0.402) 0.162 (− 0.099;0.424)

NS NS NS

0.457 (0.226;0.688) 0.445 (0.160;0.731) 0.468 (0.198;0.737)

< 0.001 0.003 < 0.001

Ref. Ref. Ref.

0.078 (− 0.157;0.313) 0.001 (− 0.283;0.283) 0.100 (− 0.171;0.371)

NS NS NS

0.427 (0.192;0.662) 0.388 (0.095–0.680) 0.434 (0.160;0.708)

0.001 0.010 0.002

Hcy = homocysteine; BPV = blood pressure variability; Ref = reference; β = standardized β-estimates. a Model 1 was adjusted for age, gender, BMI, history of diabetes mellitus, smoking status, total cholesterol, uric acid, creatinine, HbA1c, and hemoglobin. b Model 2 was additionally adjusted for the corresponding blood pressure values.

35

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

Fig. 3. Subgroup analysis based on the risk stratification of hypertension for (A.B) 24-h systolic and diastolic blood presure variability (SBPV and DBPV); (C.D) daytime SBPV and DBPV; (E.F) nighttime SBPV and DBPV. The associations between Hcy and blood presure variability seemed to be enhanced by the increased risk stratification of hypertension. Data were presented with 95% confidence intervals. P1 was the p-value for Model 1 which was adjusted for age, gender, body mass index, history of diabetes mellitus, smoking status, total cholesterol, uric acid, creatinine, HbA1c, and hemoglobin. P2 was the p-value for Model 2 which was additionally adjusted for the corresponding blood presure values. T1 = the first tertile; T2 = the second tertile; T3 = the third tertile.

reduced the risk of first stroke in hypertensive patients [5]. However, the influence of folic acid on BPV was not evaluated. Whether the ability of folic acid on stroke prevention can be partly explained by its effectiveness on lowing BPV should be further evaluated. Besides, there are still limited evidences on the ability of specific drug classes to reduce BPV, as well as their possibility to reduce the development of organ damage and the risk of cardio-cerebral vascular events [8]. Some studies showed that, compared with other antihypertensive drugs, calcium channel blockers may significantly lower BPV independently of mean blood pressure reductions, leading to a better effect on lowing the risk of stroke [10–12]. In addition, bedtime administration of the angiotensin receptor blocker was considered to not only been effective in restoring nighttime dipping of blood pressure, but also in reducing urinary albumin excretion [13]. Whether these specific drug classes remain effective on lower BPV in hypertensive patients with HHcy, unfortunately, is never evaluated. A recent study found that elevated homocysteine concentrations could decrease the antihypertensive effect of angiotensin-converting enzyme inhibitors in hypertensive patients [14]. Thus, it is also very interesting to further evaluate the interactive effect of Hcy and specific antihypertensive drug on BPV in hypertensive patients. Based on the subgroup analysis in the present study, furthermore, the associations between Hcy and ambulatory BPV seemed to be enhanced by the increased risk stratification of hypertension. Linear trends between Hcy tertiles and BPV variables were generally statistically significant for high-risk and very high-risk patients, with the only exception of nighttime DBPV. However, for patients with low-to-moderate risk, linear trends were only observed in 24-h and daytime SBPV.

tertiles showed positive linear trends with all BPV variables for very high-risk subjects. With the exception of nighttime DBPV, linear trends between the other BPV variables and Hcy tertiles were also statistically significant for high-risk patients. For those with low-to-moderate risk, however, Hcy tertiles only showed positive linear trends with 24-h and daytime SBPV. 4. Discussion We found that ambulatory BPV variables were generally increasing along with the rises in plasma Hcy concentrations. Positive and statistically significant linear trends were observed between Hcy tertiles and BPV variables. Univariate linear regression analysis indicated that Hcy tertiles were significantly associated with most of BPV variables, while multivariate analysis further confirmed that these associations were independently of the corresponding mean blood pressures and other confounding factors. Both HHcy and BPV are demonstrated to be independent predictors of stroke, and the present study confirmed the relationship between Hcy and BPV in untreated hypertensive patients [3,6]. HHcy and BPV can be easily and routinely assessed in the clinic. An early estimation of HHcy and BPV is recommended in hypertensive patients, since it may offer useful reference for the assessment of stroke risk. Furthermore, the therapeutic goal for hypertensive patients with HHcy should not merely be focused on the control of mean blood pressure values. A comprehensive antihypertensive treatment should also be targeted to lowing both HHcy and BPV. Recent study demonstrated that the combined use of folic acid and enalapril, compared with enalapril alone, significantly 36

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

This finding indicated that Hcy measurements could more reliably estimate BPV in high-risk and very high-risk populations. Furthermore, it seemed like that non-linear trends were more common for DBPV when compared to SBPV. Previous studies showed that there were more evidences for supporting the association between SBPV and future cardio-cerebrovascular events than that of DBPV, indicating the clinical importance of HHcy-associated SBPV increase [15]. In addition, when compared to daytime BPV, non-linear trends were more common for nighttime BPV. Besides, in multivariate regression analysis, negative association was observed in nighttime BPV of T2 subjects. However, there remains no evidence indicating that daytime BPV is more valid to predict adverse events than nighttime BPV. Further study on the difference of clinical significance between daytime and nighttime BPV is required. The mechanism-link between Hcy and BPV is not clear. Changes of BPV-related hormones have been observed in subjects with HHcy, such as the activation of the renin-angiotensin system, and the suppression of the nitric oxide production [1,16]. Moreover, HHcy might induce dysfunction of the cardiovascular autonomic system, an influencing factor that is considered to be related with BPV [17]. In rat modals, researchers found that moderate HHcy (12–30 μmol/l) was associated with an increase in sympathetic modulation and a decrease in parasympathetic modulation [18]. However, this association was not

observed in an early clinical study, which indicated no association between homocysteine concentrations and cardiovascular autonomic function in either diabetic or nondiabetic subjects [19]. Due to the heterogeneity of the previous studies, the direct mechanism-links between HHcy and ambulatory BPV should be further investigated in the future. In conclusion, concentration of plasma Hcy was generally and positively associated with the ambulatory BPV in patients with untreated hypertension. However, variation of evidences can be observed among different characteristics of population and diverse phenotypes of BPV. Clinical significance of HHcy-associated elevation in BPV should be further investigated in the near future. List of abbreviations HHcy Hcy BPV SBP DBP SBPV DBPV

hyperhomocysteinemia homocysteine blood pressure variability systolic blood pressure diastolic blood pressure systolic blood pressure variability diastolic blood pressure variability

Appendix A. Subgroup linear trends analysis based on the risk stratification of hypertension

BPV measurements

24-h systolic BPV Univariate Model 1a Model 2b 24-h diastolic BPV Univariate Model 1 Model 2 Daytime systolic BPV Univariate Model 1 Model 2 Daytime diastolic BPV Univariate Model 1 Model 2 Nighttime systolic BPV Univariate Model 1 Model 2 Nighttime diastolic BPV Univariate Model 1 Model 2

Low-moderate risk

High risk

Very high risk

β (95%CI)

p-value

β (95%CI)

p-value

β (95%CI)

p-value

0.401(0.060;0.473) 0.345(0.029;0.418) 0.349(0.031;0.428)

0.023 0.043 0.040

0.545(0.129;0.961) 0.497(0.113;0.857) 0.503(0.120;0.895)

0.014 0.038 0.027

0.601(0.315;0.888) 0.574(0.189;0.960) 0.614(0.255;0.972)

< 0.001 0.005 0.002

0.528(0.166;0.890) 0.745(− 1.745;3.239) 0.738(− 3.405;4.881)

0.008 NS NS

0.591(0.267;0.915) 0.524(0.072;0.977) 0.529(0.105;0.987)

0.001 0.026 0.019

0.572(0.287;0.858) 0.745(0.414;1.077) 0.725(0.417;1.034)

< 0.001 < 0.001 < 0.001

0.456(0.151;0.901) 0.430(0.109;0.909) 0.431(0.113;0.932)

0.008 0.016 0.011

0.497(0.126;0.948) 0.464(0.081;0.970) 0.468(0.083;0.972)

0.020 0.035 0.024

0.559(0.296;0.822) 0.623(0.255;0.991) 0.641(0.285;0.998)

< 0.001 0.002 0.001

0.570(0.146;0.993) 1.183(− 0.924;3.289) 1.198(− 2.283;4.679)

0.012 NS NS

0.570(0.204;0.936) 0.519(0.021;1.026) 0.520(0.004;1.044)

0.003 0.042 0.046

0.489(0.219;0.759) 0.719(0.418;1.020) 0.710(0.421;0.998)

0.001 < 0.001 < 0.001

0.413(0.030;0.795) 0.385(− 1.141;0.911) 0.388(− 0.126;0.903)

0.035 NS NS

0.487(0.161;0.933) 0.423(0.018;0.912) 0.426(0.036;0.986)

0.002 0.018 0.003

0.352(0.040;0.663) 0.447(0.001;0.895) 0.501(0.094;0.908)

0.028 0.041 0.018

0.460(0.023;0.896) 0.386(− 0.198;0.971) 0.388(− 0.184;0.960)

0.040 NS NS

0.239(− 0.185;0.662) − 0.222(−2.637;2.192) − 0.318(−5.021;4.385)

NS NS NS

0.355(0.072;0.639) 0.444(0.059:0.828) 0.463(0.116;0.809)

0.015 0.025 0.011

BPV = blood pressure variability; β = standardized β-estimates a Model 1 was adjusted for age, gender, BMI, history of diabetes mellitus, smoking status, total cholesterol, uric acid, creatinine, HbA1c, and hemoglobin. b Model 2 was additionally adjusted for the corresponding blood pressure values.

[3] Homocysteine Studies Collaboration, Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis, JAMA 288 (2002) 2015–2022. [4] H. Pang, B. Han, Q. Fu, Z. Zong, Association of high homocysteine levels with the risk stratification in hypertensive patients at risk of stroke, Clin. Ther. 38 (2016) 1184–1192. [5] Y. Huo, J. Li, X. Qin, et al., Efficacy of folic acid therapy in primary prevention of stroke among adults with hypertension in China: the CSPPT randomized clinical trial, JAMA 313 (2015) 1325–1335.

References [1] M.M. Steed, S.C. Tyagi, Mechanisms of cardiovascular remodeling in hyperhomocysteinemia, Antioxid. Redox Signal. 15 (2011) 1927–1943. [2] C. Antoniades, A.S. Antonopoulos, D. Tousoulis, K. Marinou, C. Stefanadis, Homocysteine and coronary atherosclerosis: from folate fortification to the recent clinical trials, Eur. Heart J. 30 (2009) 6–15.

37

Clinica Chimica Acta 477 (2018) 32–38

S. Li et al.

stroke, Lancet Neurol. 9 (2010) 469–480. [13] R.C. Hermida, C. Calvo, D.E. Ayala, J.E. Lopez, Decrease in urinary albumin excretion associated with the normalization of nocturnal blood pressure in hypertensive subjects, Hypertension 46 (2005) 960–968. [14] X. Qin, Y. Li, N. Sun, et al., Elevated homocysteine concentrations decrease the antihypertensive effect of angiotensin-converting enzyme inhibitors in hypertensive patients, Arterioscler. Thromb. Vasc. Biol. 37 (2017) 166–172. [15] P.M. Rothwell, S.C. Howard, E. Dolan, et al., Prognostic significance of visit-to-visit variability, maximum systolic blood pressure, and episodic hypertension, Lancet 375 (2010) 895–905. [16] U. Sen, M. Herrmann, W. Herrmann, S.C. Tyagi, Synergism between AT1 receptor and hyperhomocysteinemia during vascular remodeling, Clin. Chem. Lab. Med. 45 (2007) 1771–1776. [17] J. Conway, N. Boon, C. Davies, J.V. Jones, P. Sleight, Neural and humoral mechanisms involved in blood pressure variability, J. Hypertens. 2 (2) (1984) 203–208. [18] R.H. Mendes, C. Mostarda, G.O. Candido, et al., Moderate hyperhomocysteinemia provokes dysfunction of cardiovascular autonomic system and liver oxidative stress in rats, Auton. Neurosci. 180 (2014) 43–47. [19] A.M. Spoelstra-De Man, Y.M. Smulders, J.M. Dekker, et al., Homocysteine levels are not associated with cardiovascular autonomic function in elderly Caucasian subjects without or with type 2 diabetes mellitus: the Hoorn study, J. Intern. Med. 258 (2005) 536–543.

[6] X. Men, W. Sun, F. Fan, et al., China stroke primary prevention trial: visit-to-visit systolic blood pressure variability is an independent predictor of primary stroke in hypertensive patients, J. Am. Heart Assoc. 6 (2017) (pii: e004350). [7] M.C. Irigoyen, K. De Angelis, F. Dos Santos, D.R. Dartora, B. Rodrigues, F.M. Consolim-Colombo, Hypertension, blood pressure variability, and target organ lesion, Curr. Hypertens. Rep. 18 (2016) 31. [8] G. Parati, J.E. Ochoa, G. Bilo, Blood pressure variability, cardiovascular risk, and risk for renal disease progression, Curr. Hypertens. Rep. 14 (2012) 421–431. [9] G. Mancia, R. Fagard, K. Narkiewicz, et al., 2013 ESH/ESC guidelines for the Management of Arterial Hypertension: the task force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC), Eur. Heart J. 34 (2013) 2159–2219. [10] Y. Zhang, D. Agnoletti, M.E. Safar, J. Blacher, Effect of antihypertensive agents on blood pressure variability: the Natrilix SR versus candesartan and amlodipine in the reduction of systolic blood pressure in hypertensive patients (X-CELLENT) study, Hypertension 58 (2011) 155–160. [11] Y. Matsui, M.F. O'Rourke, S. Hoshide, J. Ishikawa, K. Shimada, K. Kario, Combined effect of angiotensin II receptor blocker and either a calcium channel blocker or diuretic on day-by-day variability of home blood pressure: the Japan combined treatment with Olmesartan and a calcium-channel blocker versus Olmesartan and diuretics randomized efficacy study, Hypertension 59 (2012) 1132–1138. [12] P.M. Rothwell, S.C. Howard, E. Dolan, et al., Effects of beta blockers and calciumchannel blockers on within-individual variability in blood pressure and risk of

38