Abstract
Ambulatory blood pressure monitoring (ABPM) is central in the management of hypertension. Factors related to BP, such as body mass index (BMI), may differently affect particular aspects of 24‐hour ABPM profiles. However, the relevance of BMI, the most used index of adiposity, has been underappreciated in the determination of specific aspects of 24‐hour ABPM profiles in hypertension. The authors evaluated the association between BMI and aspects of ABPM together with their associations with cardiac remodeling in 1841 patients. A positive association of BMI with 24‐hour, daytime, and nighttime pulse pressure in untreated normal weight and overweight/obese hypertensive patients and a positive association of BMI with nocturnal BP parameters in treated overweight/obese hypertensive patients was observed. The clinical relevance of these findings was supported by the positive significant correlations of BMI‐related BPs with left ventricular mass and atrial diameter.
Increasing adiposity is an alarming epidemic. The worldwide prevalence of obesity has nearly doubled in the past few decades, with more than 35% of adults either overweight or obese (body mass index [BMI] ≥25 kg/m2), reaching a total of more than half a billion adults worldwide.1 The prevalence of overweight/obese (OW/OB) patients is also increasing in Italy, with rates of 52.0% to 55.3% in men and 33.6% to 34.5% in women between 2001 and 2008, particularly in Southern Italy and in less educated people.2 Overweight and obesity are associated with elevated blood pressure (BP) and about 75% of cases of essential hypertension have been estimated to depend on the OW/OB phenotype.3, 4 From increasing adiposity to hypertension, the mechanisms involved often include an excess of sodium intake with many factors that increase the effect of salt on BP such as renin‐angiotensin‐aldosterone system dysregulation, cardiac natriuretic peptides system “handicap,” sympathetic nervous system overactivity, and others, leading to elevated BP values and cardiac damage.5, 6
In the several past decades, ambulatory blood pressure monitoring (ABPM) has assumed an increasing role in the management of hypertension. ABPM allows physicians to consider different aspects of the patient's BP (24‐hour, daytime, nighttime, pulse pressures (PPs), dipping or nondipping, as well as others) associated with different grades of BP‐related target organ changes and damage.
In this context, factors such as age, sex, and BMI may differently affect individual aspects of ABPM, leading to target organ damage.
However, in most, if not all, large studies on 24‐hour BP measured with ABPM, the majority of untreated and treated hypertensive patients were OW or OB. Nevertheless, the relevance of BMI, the most commonly used index of adiposity in clinical settings, has been widely underappreciated in the determination of particular aspects of 24‐hour ABPM profiles.
In 2005, Kotsis and colleagues7 studied the association of BMI with 24‐hour BP profile and found a positive correlation between BMI and many ABPM parameters in normotensive and never‐treated hypertensive patients. A few other studies have shown a reduced nocturnal fall in BP in OW/OB8 patients or higher systolic BP (SBP) and lower diastolic BP (DBP), both with office9 and ABPM measurements,10, 11 in untreated OW/OB patients, suggesting that obesity is associated with higher PP, which is one of the highest potential risks for organ damage12, 13 together with nighttime hypertension.14
The aim of our study was an in‐depth focused reevaluation of the associations between BMI and the main ABPM parameters in a large population of hypertensive patients, both normal‐weight (NW) and OW/OB, with or without antihypertensive treatment. In order to verify that the associations between BMI and ABPM parameters were not merely a “chance finding” or so small to not be clinically relevant, we analyzed the associations of BMI‐related BP parameters with left ventricular mass (LVM) and left atrial diameter (LAD), common validated indexes of afterload‐related cardiac remodeling with clinical relevance.
Patients and Methods
A retrospective study on hypertensive patients consecutively referred to our hypertension center from September 2002 to October 2011. Inclusion criteria were patients with ABPM and cardiovascular evaluation that included an echocardiography on the basis of guideline‐led clinical indications. Exclusion criteria were age younger than 18 years, low‐quality pressure monitoring (rate of artifacts >25% and/or recording duration <20 continuous hours and/or <2 recordings per hour during the day and 1 recording per hour at night), and unreliable data regarding antihypertensive treatments and secondary hypertension. Echocardiographic data were considered only if performed within 3 months from the date of the ABPM, and patients taking antihypertensive treatment were included only if on stable drug therapy for at least 6 months. Poor‐quality echocardiographic images with unreliable measures were excluded. After selection based on the above criteria, 1841 hypertensive patients were studied: 557 who did not take any antihypertensive drug (either in the initial clinical evaluation phase and/or when BP values did not require immediate drug therapy) and 1284 on antihypertensive treatment.
Body weight and height were measured on a standard beam balance scale with an attached ruler. Body weight was measured to the nearest 0.1 kg and height was measured to the nearest 1 cm. The few published studies on the association between BMI and ABPM parameters were generally in OW considered together with OB patients; therefore, we analyzed BMI both as a continuous variable and also after patient categorization as NW (BMI <25kg/m2), taken as the reference group, and as OW/OB (BMI ≥25kg/m2).
ABPM was performed using Spacelabs 90207 and 90217 (SpaceLabs Healthcare, Snoqualmie, WA), with cuff and bladder dimensions adapted to the arm circumference. Patients presented to our office with BP readings taken by other physicians. Before ABPM, office BP readings in seated patients were taken by a physician of our center on the non‐dominant arm following a 5‐minute rest in a sitting position with an A&D Medical UM‐101 (mercury‐free) device (San Jose, CA) with correct cuff size according to arm circumference. Two minutes after the first evaluation, the measurement was repeated on the other arm. In case of a difference >4 mm Hg between the two arms, a third measurement was performed on the arm in which the higher BP was measured. The arm with the higher readings was then used for ABPM. Twenty‐four–hour BP, daytime BP (defined as the BP values from 6 am to 10 pm), nighttime BP (defined as the BP values from 10 pm to 6 am), and PP (defined as SBP – DBP) were evaluated for each patient. The hours defining the “day” or “night” periods in our center were based on the most common answers to a questionnaire in which patients were asked about their sleeping behavior. The night‐to‐day ratio, as described,15 was the ratio between mean nighttime and daytime ABPM values. Night‐to‐day ratios were multiplied by 100, therefore expressing nighttime BP as a percentage of a daytime level. A ratio of 100% or higher signified the absence of a BP fall at night. We considered patients whose mean nighttime SBP was at least 10% below the mean daytime values as dippers. Among patients taking antihypertensive treatment, those with mean 24‐hour BP <130/80 mm Hg, mean daytime BP <135/85 mm Hg, and mean nighttime BP <120/70 mm Hg were defined as controlled hypertensives.16
LVM and LAD are considered “integrated indexes” of pressure afterload and are the most used and validated measures in clinical practice. Left ventricular dimensions were measured by echocardiography (ATL HDI 5000; Philips Medical, Amsterdam, The Netherlands) and calculation of LVM was according to recommendations from the European Society of Hypertension (ESH).16 LVM was indexed by body surface area for NW and by height2.7 and height1.7 for OW/OB as indicated by the recent ESH/European Society of Cardiology hypertension 2013 guidelines.16
A treatment intensity score (TIS) was calculated to allow comparison of drug regimens across patients taking many different combinations of medications. As previously reported,17 the daily dose taken recorded by the patient was divided by the maximum recommended daily dose to obtain a proportional dose (called “intensity”) for that medication. For completeness, dual‐class drugs were separated into their components, and intensity was calculated separately for each chemical compound. The maximum recommended daily doses set by the Italian National Drug Agency (AIFA) were used for calculations. The sum of all the different values was recorded as TIS and was used for adjusting results.
Statistical Analysis
Data were analyzed with the Statistical Package for Social Science version 13 (SPSS Inc, Chicago, IL). A value of P<.05 was defined as statistically significant. Normally distributed continuous variables are expressed as mean±standard deviation. Categorical variables are expressed as absolute number and percentage. Pearson correlation was used to analyze the relationship between continuous variables, and multiple linear regression was used for adjusting for covariates. The χ² test was used to analyze the differences between categorical variables, and logistic regression was used to adjust results.
Results
General characteristics of 557 untreated and 1284 treated patients are shown in Table 1.
Table 1.
General Characteristics of Untreated and Treated Patients
| Untreated (557) | Treated (1284) | P Value | |
|---|---|---|---|
| Age, y | 50.6±13.9 | 59.0±12.3 | <.001 |
| Sex | |||
| Male, % | 344 (61.8) | 715 (55.7) | .015 |
| Female, % | 213 (38.2) | 569 (44.3) | |
| BMI | 26.5±4.2 | 27.7±4.6 | <.001 |
| 24‐h SBP | 137.2±12.3 | 131.2±14.2 | <.001 |
| 24‐h DBP | 85.5±9.6 | 78.3±10.4 | <.001 |
| Daytime SBP | 140.3±12.6 | 134.3±14.5 | <.001 |
| Daytime DBP | 88.6±10.0 | 81.2±10.8 | <.001 |
| Nighttime SBP | 129.0±14.0 | 123.7±15.5 | <.001 |
| Nighttime DBP | 77.4±10.3 | 71.3±10.6 | <.001 |
| 24‐h PP | 51.7±10.0 | 52.9±11.2 | .023 |
| Daytime PP | 51.7±10.0 | 53.1±11.4 | .008 |
| Nighttime PP | 51.6±10.6 | 52.4±11.7 | .130 |
| TIS | 1.23±0.75 | ||
| LVM | 205.8±59.9 | 220.0±64.9 | <.001 |
| LVM/BSA | 107.7±27.4 | 114.9±28.6 | <.001 |
| LVM/h2.7 | 49.0±13.6 | 54.3±15.0 | <.001 |
| LVM/h1.7 | 83.1±22.6 | 90.9±24.7 | <.001 |
| LAD | 37.5±5.0 | 38.8±5.5 | <.001 |
Abbreviations: BMI, body mass index; BSA, body surface area; DBP, diastolic blood pressure; LAD, left atrial diameter; LVM/h1.7, left ventricular mass (LVM) by height; PP, pulse pressure; SBP, systolic blood pressure; TIS, treatment intensity score. All continuous variables are expressed as mean±standard deviation. All categorical variables are expressed as number (percentage).
BMI correlated with 24‐hour (r=0.105; P=.013), daytime (r=0.097; P=.022) and nighttime SBP (r=0.115; P=.007), and PP in untreated hypertensive patients, with better correlations with PPs (r=0.180, r=0.173, r=0.178, respectively, for 24‐hour, daytime, nighttime PP; all P<.001). Associations were confirmed after adjusting for age and sex in multiple linear regression models (Table 2). In treated hypertensive patients, BMI correlated only with nighttime SBP (r=0.090; P=.001), nighttime DBP (r=0.059; P=.035), and nighttime PP (r=0.065; P=.020) and with the night‐to‐day ratio (r=0.158; P<.001). Associations between BMI and nighttime BP findings were significant in both controlled and uncontrolled patients when analyzed separately (data not shown). After adjusting for age, sex, TIS, and BP control, associations were maintained for nighttime SBP, PP, and night‐to‐day ratio (Table 2).
Table 2.
Correlations Between BMI and ABPM Parameters After Adjusting for Age and Sex in Untreated Patients and for Age, Sex, TIS, and BP Control in Treated Patients
| Change in BP (95% CI) | Beta | P Value | |
|---|---|---|---|
| Untreated | |||
| 24‐h PP | 0.40 (0.21/0.60) | 0.169 | <.001 |
| Daytime PP | 0.40 (0.20/0.60) | 0.151 | <.001 |
| Nighttime PP | 0.43 (0.22/0.63) | 0.167 | <.001 |
| Treated | |||
| Nighttime SBP | 0.31 (0.16/0.46) | 0.093 | <.001 |
| Nighttime DBP | −0.03 (−0.7/0.13) | −0.005 | ns |
| Nighttime PP | 0.28 (0.16/0.40) | 0.111 | <.001 |
| Night‐to‐day ratio | 0.25 (0.17/0.32) | 0.168 | <.001 |
Abbreviations: ABPM, ambulatory blood pressure monitoring; CI, confidence interval; DBP, diastolic blood pressure; ns, not significant; PP, pulse pressure; SBP, systolic blood pressure. Data were adjusted for age and sex (for untreated) and for age, sex, treatment intensity score (TIS), and blood pressure (BP) control (for treated) in multiple linear regression models. Change in BP represents the amount of change in BPs in mm Hg (on the average) from a one‐unit increase in body mass index (BMI) (kg/m2). Bold values indicate significance.
For a further understanding of the relationship between BMI and these specific aspects of ABPM data, correlations of BMI with BP parameters, and their associations with cardiac remodeling were analyzed separately in untreated and treated NW and OW/OB hypertensive patients, as shown below.
Untreated Hypertensive Patients
Of the 557 hypertensive patients who were not taking any antihypertensive drug, 215 (38.6%) were NW and 342 (61.4%) were OW/OB (18% had a BMI ≥30kg/m2). The most important finding in the untreated population was that BMI was positively correlated with 24‐hour PP, daytime PP, and nighttime PP. These correlations were confirmed both in untreated NW and OW/OB patients after adjusting for age and sex in multiple linear regression models (Table 3).
Table 3.
Correlations Between BMI and ABPM Parameters in Untreated NW and OW/OB Hypertensive Patients (Adjusted for Age and Sex)
| Change in BP (95% CI) | Beta | P Value | |
|---|---|---|---|
| NW | |||
| 24‐h SBP | −0.17 (−1.14/0.80) | −0.026 | ns |
| 24‐h DBP | −0.93 (−1.71/−0.15) | −0.166 | .020 |
| Daytime SBP | −0.21 (−1.20/0.78) | −0.32 | ns |
| Daytime DBP | −1.01 (−1.81/−0.20) | −0.173 | .015 |
| Nighttime SBP | −0.08 (−1.20/1.03) | −0.01 | ns |
| Nighttime DBP | −0.70 (−1.54/0.15) | −0.120 | ns |
| 24‐h PP | 0.77 (0.02/1.51) | 0.145 | .045 |
| Daytime PP | 0.79 (0.05/1.54) | 0.151 | .037 |
| Nighttime PP | 0.61 (−0.19/1.42) | 0.109 | ns |
| OW/OB | |||
| 24‐h SBP | 0.50 (0.12/0.90) | 0.138 | .012 |
| 24‐h DBP | −0.19 (−0.30/0.26) | −0.007 | ns |
| Daytime SBP | 0.49 (0.10/0.90) | 0.131 | .018 |
| Daytime DBP | −0.02 (−0.30/0.26) | −0.007 | ns |
| Nighttime SBP | 0.64 (0.19/1.08) | 0.153 | .005 |
| Nighttime DBP | 0.04 (−0.28/0.35) | 0.012 | ns |
| 24‐h PP | 0.52 (0.21/0.83) | 0.176 | .001 |
| Daytime PP | 0.51 (0.20/0.83) | 0.170 | .002 |
| Nighttime PP | 0.60 (0.27/0.93) | 0.190 | <.001 |
Abbreviations: ABPM, ambulatory blood pressure monitoring; CI, confidence interval; DBP, diastolic blood pressure; ns, not significant; NW, normal‐weight; OW/OB, overweight/obese; PP, pulse pressure; SBP, systolic blood pressure. Data were adjusted for age and sex in multiple linear regression models. Change in blood pressure (BP) represents the amount of change in each BP in mm Hg (on the average) from a one‐unit increase in body mass index (BMI) (kg/m2). Bold values indicate significance.
In NW patients, this association appeared to be mainly driven by a negative correlation of BMI with 24‐hour DBP, daytime DBP, and nighttime DBP, whereas in OW/OB patients there appeared to be a positive correlation of BMI with 24‐hour, daytime, and nighttime SBP (Table 3).
No correlation emerged between BMI and night‐to‐day ratio and no difference in prevalence of nondippers between NW and OW/OB patients.
BMI‐Related PPs and Cardiac Remodeling in Untreated Patients
To verify that the BMI‐related BP parameters in untreated patients were meaningful, we correlated the findings with indexes of cardiac remodeling/afterload. All BMI‐related PPs were positively associated with LVM and LAD both in NW and OW/OB patients. Associations were maintained after adjusting for sex, age, and BMI in multiple linear regression models (Table 4).
Table 4.
Correlations of BMI‐Related PPs With LVM (Differently Indexed for NW and OW/OB) and LAD
| Change in LVM and LAD (95% CI) | Beta | P Value | |
|---|---|---|---|
| NW patients | |||
| LVM | |||
| 24‐h PP | 1.70 (1.00/2.39) | 0.274 | <.001 |
| Daytime PP | 1.65 (0.96/2.34) | 0.267 | <.001 |
| LVM/BSA | |||
| 24‐h PP | 0.96 (0.56/1.36) | 0.296 | <.001 |
| Daytime PP | 0.94 (0.54/1.34) | 0.291 | <.001 |
| LAD | |||
| 24‐h PP | 0.10 (0.03/0.16) | 0.182 | .004 |
| Daytime PP | 0.10 (0.04/0.17) | 0.189 | .003 |
| OW/OB patients | |||
| LVM | |||
| 24‐h PP | 1.03 (0.51/1.55) | 0.188 | <.001 |
| Daytime PP | 0.98 (0.47/1.49) | 0.182 | <.001 |
| Nighttime PP | 0.88 (0.38/1.37) | 0.170 | .001 |
| LVM/h2.7 | |||
| 24‐h PP | 0.29 (0.16/0.41) | 0.227 | <.001 |
| Daytime PP | 0.28 (0.15/0.40) | 0.223 | <.001 |
| Nighttime PP | 0.25 (0.13/0.36) | 0.209 | <.001 |
| LVM/h1.7 | |||
| 24‐h PP | 0.46 (0.25/0.66) | 0.221 | <.001 |
| Daytime PP | 0.44 (0.24/0.64) | 0.216 | <.001 |
| Nighttime PP | 0.39 (0.20/0.59) | 0.202 | <.001 |
| LAD | |||
| 24‐h PP | 0.07 (0.02/0.11) | 0.156 | .004 |
| Daytime PP | 0.06 (0.02/0.11) | 0.148 | .006 |
| Nighttime PP | 0.07 (0.02/0.11) | 0.166 | .002 |
Abbreviations: CI, confidence interval; NW, normal‐weight; OW/OB, overweight/obese. Data were adjusted for age, sex, and body mass index (BMI) in multiple linear regression models. Change in left ventricular mass (LVM) and left atrial diameter (LAD) represents the amount of change in each dependent variable (LVM, LVM/body surface area [BSA], LVM by height2.7 [LVM/h2.7], LVM by height1.7 [LVM/h1.7], and LAD) from a one‐unit increase in pulse pressure (PP) (24‐hour, daytime, nighttime). Bold values indicate significance.
Treated Hypertensive Patients
Of the 1284 hypertensive patients with antihypertensive treatment, 344 (26.8%) were controlled, whereas 940 (73.2%) were not controlled at ABPM. The most important finding in the treated population was that BMI was positively correlated with nighttime SBP and PP. These correlations were confirmed in treated OW/OB patients (903, 70.3%) after adjusting for age, sex, TIS, and BP control in multiple linear regression models (Table 5). Noteworthy, BMI did not show any correlation with BP parameters in treated NW patients.
Table 5.
Correlations Between BMI and Nighttime SBP, PP, and Night‐to‐Day Ratio in Treated OW/OB Hypertensive Patients
| Change in BPs and Night‐to‐Day Ratio (95% Confidence Interval) | Beta | P Values | |
|---|---|---|---|
| OW/OB | |||
| Nighttime SBP | 0.48 (0.27/0.69) | 0.118 | <.001 |
| Nighttime PP | 0.42 (0.25/0.60) | 0.142 | <.001 |
| Night‐to‐day ratio | 0.32 (0.21/0.42) | 0.193 | <.001 |
Data were adjusted for age, sex, treatment intensity score, and blood pressure (BP) control in multiple linear regression models. Change in BPs and in night‐to‐day ratio represents the amount of change in nighttime systolic BP (SBP), nighttime pulse pressure (PP), and night‐to‐day ratio (on the average) from a one‐unit increase in body mass index (BMI) (kg/m2). Bold values indicate significance.
The association between BMI and nighttime BP translates into an altered circadian rhythm of BP in treated OW/OB patients with a positive correlation between BMI and night‐to‐day ratio and a higher prevalence of nondipping in OW/OB patients (54.0% vs 43.3% in NW patients; odds ratio, 1.57; P=.001) as shown in the Figure (panel A). These results were confirmed after adjusting for age, sex, TIS, and BP control in multiple linear regression models and logistic regression (Table 5 and Figure [panel A]).
Figure 1.
Panel A: A higher prevalence of nondipping was found in overweight/obese (OW/OB) patients vs normal‐weight (NW) patients. Panel B: A higher prevalence of left ventricular hypertrophy (LVH) was found in nondippers vs dippers. OR indicates odds ratio; CI, confidence interval.
BMI‐Related Nighttime BP and Cardiac Remodeling in Treated Patients
Nighttime SBP and PP and night‐to‐day ratio were positively associated with LVM and LAD. The main associations were maintained after adjusting for sex, age, BMI, TIS, and BP control in multiple linear regression models (Table 6). The Figure (panel B) shows the higher prevalence of left ventricular hypertrophy (LVH) in nondippers vs dippers (defined on the basis of the LVM/h2.7 using ≥49.2 g/m2.7 in men and ≥46.7g/m2.7 in women as partition values18).
Table 6.
Correlations of BMI‐Related Nocturnal BP Parameters With LVM (and LVM Indexed for OW/OB) and LAD
| Change in LVM and LAD (95% Confidence Interval) | Beta | P Value | |
|---|---|---|---|
| OW/OB | |||
| LVM | |||
| Nighttime SBP | 0.75 (0.45/1.05) | 0.182 | <.001 |
| Nighttime PP | 0.58 (0.20/0.95) | 0.104 | .003 |
| Night‐to‐day ratio | 0.81 (0.20/1.42) | 0.081 | .010 |
| LVM/h2.7 | |||
| Nighttime SBP | 0.18 (0.11/0.26) | 0.196 | .000 |
| Nighttime PP | 0.15 (0.06/0.24) | 0.118 | .001 |
| Night‐to‐day ratio | 0.14 (−0.01/0.29) | 0.061 | ns |
| LVM/h1.7 | |||
| Nighttime SBP | 0.31 (0.19/0.42) | 0.199 | <.001 |
| Nighttime PP | 0.24 (0.09/0.39) | 0.116 | .001 |
| Night‐to‐day ratio | 0.27 (0.03/0.51) | 0.072 | .027 |
| LAD | |||
| Nighttime SBP | 0.03 (0.01/0.06) | 0.090 | .029 |
| Nighttime PP | 0.07 (0.03/0.10) | 0.147 | <.001 |
| Night‐to‐day ratio | 0.03 (−0.03/0.08) | 0.032 | ns |
Abbreviations: ns, not significant; OW/OB, overweight/obese. Data were adjusted for age, sex, body mass index (BMI), treatment intensity score, and blood pressure (BP) control in multiple linear regression models. Change in left ventricular mass (LVM) and left atrial diameter (LAD) represents the amount of change in every dependent variable (LVM, LVM by height2.7 [LVM/h2.7], LVM by height1.7 [LVM/h1.7], and LAD) from a one‐unit increase in nighttime systolic BP (SBP), pulse pressure (PP), and night‐to‐day ratio. Bold values indicate significance.
Discussion
To evaluate in‐depth the associations between BMI and specific ABPM parameters and the relevance of the BMI‐related pressure data on cardiac remodeling we studied a large population with clinical indications for ABPM and cardiovascular evaluation that included echocardiography.
Associations Between BMI, BP Parameters, and Cardiac Remodeling in Untreated Hypertensive Patients
We found a positive association between BMI and 24‐hour, daytime, and nighttime PP in the group of untreated hypertensive patients (Table 2). The relationship between BMI and its related BPs was quantified as an increase in 0.4 mm Hg for each unit of increase of BMI in kg/m2. An unexpected new finding, to the best of our knowledge, was the association between BMI and 24‐hour, daytime, and nighttime PP not only in OW/OB patients but also in NW hypertensive patients (Table 3). In NW patients, it was the reduction of DBP with increasing BMI that appeared to lead to the association between BMI and PP. On the contrary, in OW/OB patients, the association between BMI and PP appeared to be mainly driven by the increase in SBP with increasing BMI (Table 3). This finding might be reasonably explained by some hemodynamic features of OW/OB patients such as volume overload with consequent SBP increase.5 In hypertensive patients in the NW range (<25 kg/m2), the increase in BMI is not known to be coupled with the sodium‐retentive effect that promotes volume overload typical of OW/OB patients, but other mechanisms may result in larger peripheral vasodilatation and lower DBP. It is also possible that it is the “underweight” of some patients that drove a significant negative correlation.
The importance and clinical significance of these findings are enlightened by the positive association between BMI‐related PPs and cardiac remodeling (increased LVM and LAD) in both NW and OW/OB patients, even after adjusting for BMI itself and for two other important determinants such as age and sex (Table 4).
Therefore, the BMI‐associated PPs contribute significantly to the increased cardiac afterload and to the development of subclinical or clinically significant cardiac organ remodeling in untreated NW and OW/OB hypertensive patients.
Associations Between BMI, BP Parameters, and Cardiac Remodeling in Treated Hypertensive Patients
In treated hypertensive patients, no association between BMI and 24‐hour and daytime PPs was found. This may be reasonably explained by the fact that most antihypertensive drugs have a numerically greater effect on SBP than on DBP, leading to a “flattening” of differential BP. However, in treated OW/OB hypertensive patients, BMI affects BP, with a positive correlation with nocturnal BP parameters and with a more prevalent nondipping pattern (Table 2 and Figure [panel A]). The correlation is maintained even after adjusting for age, sex, BP control, and treatment intensity (Table 5 and Figure [panel A]). This finding should be interpreted in the light of the general characteristics of our treated population with an average TIS of 1.23. Since more than 60% of patients took at least two drugs, it means that most patients assumed the largest part of the drugs not at full dosages. Considering the higher prevalence of morning administration (94%), the lower dosages usually translate into reduced coverage after 24 hours, particularly in OW/OB patients in whom increased adiposity is associated with increased extracellular volume and relative resistance to therapy and in whom other causes of elevated nocturnal BP can be present (obstructive sleep apnea syndrome and/or greater extravascular liquid reabsorption during nighttime). Indeed, in NW patients generally more responsive to drug therapy, there were no associations between nocturnal BP and increasing BMI. Regarding the relationship of BMI with nighttime BPs, we subanalyzed the treated patients on the basis of the different drug classes (70.2% angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers, 44.2% diuretics, 43.2% calcium antagonists, and 31.5% β‐blockers) and even in the smaller subgroup (β‐blockers) the associations with BMI remained significant. The association between BMI and nighttime BP is important as it is only detectable using ABPM, underlying the importance of 24‐hour BP monitoring to verify the efficacy of prescribed therapies, especially in patients with BMI ≥25kg/m2, even when apparently controlled according to daytime BP measurements.
The relationship between BMI and its related BPs in treated patients were quantified as an increase of 0.3 mm Hg for each unit of increase of BMI in kg/m2, a value that reached 0.48 mm Hg for nighttime SBP in treated OW/OB patients (Tables 2 and 5). This finding also reinforces the importance of proper use of drugs with appropriate dosages in hypertensive patients and/or adding drugs before bedtime, especially when patients are OW/OB.
The relevance and clinical significance of our overall findings in treated patients was confirmed, as underlined above for untreated patients, by the association of BMI‐related nighttime BP parameters with cardiac remodeling, even after adjusting for BMI, age, sex, BP control, and intensity of drug treatment (Table 6 and Figure [panel B]).
Study Strengths and Limitations
The main limitation of our study is in the design as a post‐hoc analysis. However, patients were real‐life consecutive outpatients, not treated or on stable therapy, which allowed better evaluation of real BP load. ABPM recordings were performed for guideline‐led clinical reasons and patients were selected for the presence of an echocardiography performed time‐closely to the ABPM. However, we believe that our findings are remarkable because very few data are presently available on the detailed relationship between BMI and ABPM parameters supported by BP‐related changes, and, to the best of our knowledge, some of our findings are new (eg, the influence of BMI on BP in untreated NW and in OW/OB treated patients). Another relative limitation was that we did not have a complete set of echocardiographic measurements such as parameters of diastolic function. Nevertheless, the strongest studied relationships among specific ABPM datasets and the heart are mainly LVM and atrial size, and the main focus of our study was to analyze BMI‐related BPs using only the cardiac parameters to corroborate the clinical relevance of our BMI‐related BP findings. Our findings may also have important clinical implications because of the real‐life patients studied.
Conclusions
BMI is associated with particular ABPM parameters in untreated and treated hypertensive patients. The correlations between BMI‐related BP parameters with cardiac remodeling in any considered subgroup strongly support the importance and quality of these data. Therefore, not only BMI affects some BP parameters, but the BMI‐related BP parameters are likely to be the main mechanism of the BMI‐related cardiac changes.6 Specific aspects of BMI‐related ABPM parameters have been underappreciated in most studies and thus we believe that our findings will be useful for physicians who manage OW/OB hypertensive patients in the reevaluation of these clinically relevant aspects.
Disclosure
The authors report no specific funding in relation to this research and no conflicts of interest to disclose.
J Clin Hypertens (Greenwich). 2015;17:147–153. DOI: 10.1111/jch.12463. © 2015 Wiley Periodicals, Inc.
References
- 1. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980‐2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384:766–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Gallus S, Odone A, Lugo A, et al. Overweight and obesity prevalence and determinants in Italy: an update to 2010. Eur J Nutr. 2013;52:677–685. [DOI] [PubMed] [Google Scholar]
- 3. Garrison RJ, Kannel WB, Stokes J, Castelli WP. Incidence and precursors of hypertension in young adults: the Framingham Offspring Study. Prev Med. 1987;16:235–251. [DOI] [PubMed] [Google Scholar]
- 4. Brown CD, Higgins M, Donato KA, et al. Body mass index and the prevalence of hypertension and dyslipidemia. Obes Res. 2000;8:605–619. [DOI] [PubMed] [Google Scholar]
- 5. Sarzani R, Salvi F, Dessì‐Fulgheri P, Rappelli A. Renin‐angiotensin system, natriuretic peptides, obesity, metabolic syndrome, and hypertension: an integrated view in humans. J Hypertens. 2008;26:831–843. [DOI] [PubMed] [Google Scholar]
- 6. Sarzani R, Bordicchia M, Spannella F, et al. Hypertensive heart disease and obesity: a complex interaction between hemodynamic and not hemodynamic factors. High Blood Press Cardiovasc Prev. 2014;21:81–87. [DOI] [PubMed] [Google Scholar]
- 7. Kotsis V, Stabouli S, Bouldin M, et al. Impact of obesity on 24‐hour ambulatory blood pressure and hypertension. Hypertension. 2005;45:602–607. [DOI] [PubMed] [Google Scholar]
- 8. Cuspidi C, Meani S, Valerio C, et al. Body mass index, nocturnal fall in blood pressure and organ damage in untreated essential hypertensive patients. Blood Press Monit. 2008;13:318–324. [DOI] [PubMed] [Google Scholar]
- 9. Kwagyan J, Tabe CE, Xu S, et al. The impact of body mass index on pulse pressure in obesity. J Hypertens. 2005;23:619–624. [DOI] [PubMed] [Google Scholar]
- 10. Randall OS, Kwagyan J, Huang Z, et al. Effect of diet and exercise on pulse pressure and cardiac function in morbid obesity: analysis of 24‐hour ambulatory blood pressure. J Clin Hypertens (Greenwich). 2005;7:455–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. De PG, Nardecchia A, Guida P, Silvestris F. Arterial hypertension in obesity: relationships with hormone and anthropometric parameters. Eur J Cardiovasc Prev Rehabil. 2011;18:240–247. [DOI] [PubMed] [Google Scholar]
- 12. Verdecchia P, Schillaci G, Borgioni C, et al. Ambulatory pulse pressure: a potent predictor of total cardiovascular risk in hypertension. Hypertension. 1998;32:983–988. [DOI] [PubMed] [Google Scholar]
- 13. Kikuya M, Staessen JA, Ohkubo T, et al. Ambulatory arterial stiffness index and 24‐hour ambulatory pulse pressure as predictors of mortality in Ohasama, Japan. Stroke. 2007;38:1161–1166. [DOI] [PubMed] [Google Scholar]
- 14. Hansen TW, Li Y, Boggia J, et al. Predictive role of the nighttime blood pressure. Hypertension. 2011;57:3–10. [DOI] [PubMed] [Google Scholar]
- 15. Staessen JA, Bieniaszewski L, O'Brien E, et al. Nocturnal blood pressure fall on ambulatory monitoring in a large international database. The “Ad Hoc” Working Group. Hypertension. 1997;29:30–39. [DOI] [PubMed] [Google Scholar]
- 16. Mancia G, Fagard R, Narkiewicz K, 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). J Hypertens. 2013;31:1281–1357. [DOI] [PubMed] [Google Scholar]
- 17. Bailey KR, Grossardt BR, Graves JW. Novel use of Kaplan‐Meier methods to explain age and gender differences in hypertension control rates. Hypertension. 2008;51:841–847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Gerdts E, Okin PM, De Simone G, et al. Gender differences in left ventricular structure and function during antihypertensive treatment: the Losartan Intervention for Endpoint Reduction in Hypertension Study. Hypertension. 2008;51:1109–1114. [DOI] [PubMed] [Google Scholar]