Long-Term Improvement in Aortic Pulse Wave Velocity After Weight Loss Can Be Predicted by White Adipose Tissue Factors

Long-Term Improvement in Aortic Pulse Wave Velocity After Weight Loss Can Be Predicted by White... Abstract BACKGROUND Arterial stiffness, measured by pulse wave velocity (PWV), is linked to obesity, cardiovascular disease, and all-cause mortality. Short-term weight loss improves PWV, but the long-term effects are unknown. We investigated the effect of pronounced long-term weight loss on PWV and whether anthropometric/metabolic parameters and/or white adipose tissue (WAT) phenotype could predict this change in PWV. METHODS Eighty-two obese subjects were examined before and 2 years after Roux-en-Y gastric bypass. Analyses included anthropometrics, routine clinical chemistry, and hyperinsulinemic-euglycemic clamp. Arterial stiffness was measured as aortic PWV (aPWV) using the Arteriograph device. WAT mass and distribution were assessed by dual-X-ray absorptiometry. Baseline visceral and subcutaneous WAT samples were obtained to measure adipocyte cell size. Transcriptomic profiling of subcutaneous WAT was performed in a subset of subjects (n = 30). RESULTS At the 2-year follow-up, there were significant decreases in body mass index (39.4 ± 3.5 kg/m2 vs. 26.6 ± 3.4 kg/m2; P < 0.0001) and aPWV (7.8 ± 1.5 m/s vs. 7.2 ± 1.4 m/s; P = 0.006). Multiple regression analyses showed that baseline subcutaneous adipocyte volume was associated with a reduction in aPWV (P = 0.014), after adjusting for confounders. Expression analyses of 52 genes implicated in arterial stiffness showed that only one, COL4A1, independently predicted improvements in aPWV after adjusting for confounders (P = 0.006). CONCLUSIONS Bariatric surgery leads to long-term reduction in aPWV. This improvement can be independently predicted by subcutaneous adipocyte volume and WAT COL4A1 expression, which suggests that subcutaneous WAT has a role in regulating aPWV. CLINICAL TRIALS REGISTRATION Trial Number NCT01727245 (clinicaltrials.gov) adipocyte/metabolism, bariatric surgery, blood pressure, cell size, hypertension, humans, longitudinal studies, morbid/complications, obesity, vascular stiffness Arterial stiffness, a measure of reduced arterial compliance following changes in intraluminal pressure, can be measured noninvasively by pulse wave velocity (PWV). PWV is a strong predictor of future cardiovascular events and all-cause mortality,1 independent of other risk factors.2 Age, heart rate, and blood pressure explain much of the interindividual variations in arterial stiffness,3 although physical activity and smoking also contribute.4,5 Body weight gain increases arterial stiffness.6 Therefore, numerous studies have explored the relationship between obesity and arterial stiffness. Cross-sectional studies, have reported positive associations between arterial stiffness and insulin resistance,7 adipokine secretion from white adipose tissue (WAT),8 WAT distribution,9 inflammation,10 and circulating levels of free fatty acids.11 Moreover, observational studies have identified associations between carotid–femoral PWV and Hemoglobin A1c (HbA1c) levels,12 and between aortic PWV (aPWV) and central obesity, plasma adiponectin, and triglyceride levels.13 We reported that visceral adipocyte volume is significantly associated with aPWV in obese individuals.14 Altogether, these findings suggest a possible causal relationship between WAT dysfunction and arterial stiffness. Genome-wide and candidate gene-association studies have identified numerous genes whose expression associates with arterial stiffness.15 These encode proteins in the renin–angiotensin–aldosterone pathway, extracellular matrix components, metalloproteinases, proinflammatory factors, β-adrenergic and endothelin signaling receptors, endothelial cell apoptosis, and vascular wall immune response. Although most of these are expressed in peripheral tissues, including WAT, it remains unclear whether expression of any of these genes in WAT is associated with arterial stiffness. There is accumulating evidence that short-term moderate weight reduction, by means of energy-restricted diet or energy-restricted diet combined with exercise in subjects followed for up to a year, improves arterial stiffness, as measured by PWV.16 However, studies in obese subjects undergoing bariatric surgery, which causes greater weight loss than diet and exercise, have reported divergent results. For example, 2 studies showed reduced PWV,17,18 while another suggested no improvement.19 Importantly, participants were followed for only a moderate time (up to 1 year), with subjects likely not weight stable at time of examinations. To our knowledge, no studies have examined the long-term effects of weight reduction on PWV. Herein, we aimed to determine whether weight loss results in a long-term reduction in aPWV. We also aimed to explore whether baseline WAT phenotype, including anthropometric/metabolic parameters or the expression of selected genes, could predict improvements in PWV. METHODS Subjects One-hundred-and-twenty obese individuals, scheduled for bariatric surgery, were consecutively recruited in a longitudinal study of the effects of weight loss on WAT function (clinicaltrials.gov, NCT01727245). The inclusion/exclusion criteria have been described elsewhere.14 Subjects underwent bariatric surgery with Roux-en-Y gastric bypass. Given that blood pressure has a major effect on arterial stiffness, subjects using antihypertensive medication (n = 37) were excluded. One patient was excluded because of a faulty aPWV measurement. Finally, 82 subjects (71 women and 11 men) were included. The study was approved by the regional ethics board in Stockholm and was conducted in accordance with the statutes of the Declaration of Helsinki. Written informed consent was obtained from all participants. Subjects were examined before and 24 months after surgery, when they had attained a stable body weight. Subjects came to the laboratory at 7.30 a.m. after an overnight fast and having abstained from nicotine and caffeine. Data on nicotine use, medications, self-assessed physical exercise (graded in predetermined increments on a scale from 1–4), and past medical history were collected in questionnaires and confirmed by interview. Standard anthropometric measures, including body weight, height, body mass index (BMI), and waist and hip circumferences, were recorded. Framingham risk scores were calculated according to Adult Treatment Panel III criteria.20 The same staff performed all measurements and laboratory examinations both before and 2 years after surgery. Arterial stiffness and blood pressure Recommended standardized procedures for measuring arterial stiffness were followed21 and were performed exactly as described.14 Resting heart rate was recorded after 15 minutes of rest in the supine position. Aortic PWV and blood pressure was measured using an Arteriograph (TensioMed, Budapest, Hungary) with appropriate cuff size based on subjects arm circumference at the time of examination. It was calculated by dividing the distance traveled by the pulse wave by its transit time (RT/2). Mean values based on at least 3 measurements were used. The distance traveled from the suprasternal notch to the pubic bone was measured in a straight line to avoid the influence of variations in abdominal circumference. As there is no evidence that aortic length is influenced by weight loss, the measure obtained at baseline was used for both baseline and 2-year aPWV calculations. The Arteriograph also provides other vascular measures; augmentation index (the % of central pulse pressure attributed to the reflected wave from the periphery) (AIX), central systolic blood pressure (the systolic blood pressure in the aorta) (SBPao), and pulse pressure (the difference between systolic and diastolic blood pressure) (PP). Systolic and diastolic blood pressures were also obtained using an automatic device (Omron M10-IT; Omron Health Care, Hoofddorp, The Netherlands). Assessment of in vivo insulin sensitivity and body composition Subjects underwent a hyperinsulinemic-euglycemic clamp and Dual-energy X-ray absorptiometry (DXA) exactly as previously described.14 All subjects were examined using a GE Lunar iDXA running the enCORE software (version 14.10.022) with the CoreScan feature (GE Medical Systems, Chalfont St. Giles, UK). This allowed determination of body fat composition (android, gynoid, estimated visceral adipose tissue; EVAT), and estimated subcutaneous adipose tissue (ESAT) in the android region as discussed previously.14 Of note, EVAT and ESAT do not represent total abdominal visceral and subcutaneous WAT, but only the region that corresponds to where the fat biopsy is obtained. Adipocyte cell volume Adipose tissue samples were treated with collagenase to isolate adipocytes. The diameter of 100 adipocytes was determined microscopically, and the mean adipocyte cell volume was calculated as described by Hirsch and Gallian.22 This method has been used in our laboratory for over 4 decades, and the validity has been described in detail elsewhere.23 Global transcriptomics of WAT Global transcriptomic profiling was performed using 5′cap analysis of gene expression (CAGE) of subcutaneous WAT from a subset of 30 participants at the baseline, as described previously in detail.24 In brief, total RNA was extracted and the quality was determined using the bioanalyzer RNA 6000 Pico Kit (Agilent Technologies). Messenger RNA levels are expressed as tag per million normalized read counts. All raw data are uploaded at https://export.uppmax.uu.se/b2013047/CellReportsTables/. Gene expression data were analyzed in a 2-step approach. First, a global analysis accepting a false discovery rate of 0.1 was performed. Because no significant associations were found, further analysis was undertaken by examining a previously used list of 52 genes found to be associated with arterial stiffness in genetic variant studies,25 which included candidate gene studies and genome-wide association studies.15,26,27 Statistics The data are expressed as mean ± SD. The Shapiro–Wilks test was used to identify normal distributions. Normally distributed values were compared using a paired-sample Student t test (2-tailed), and non-normally distributed values were compared using the related-samples Wilcoxon signed–rank test. Any missing data were excluded on a pairwise basis. Associations between baseline or changes in (i.e., delta; Δ) WAT phenotype/metabolic parameters with ΔaPWV were tested by simple linear regression analysis, where ΔaPWV was set as the dependent variable. Bonferroni correction was applied to correct for multiple testing resulting in alpha levels of 0.05/15 = 0.0033 (baseline WAT and metabolic factors vs. ΔaPWV) and 0.05/40 = 0.00125 (WAT gene expression and ΔaPWV). All parameters that were significantly associated in the simple regression were entered into a multiple regression analysis model adjusted for systolic blood pressure and resting heart rate (Model 1). In an additional analysis, we expanded Model 1 and included other factors known to associate with PWV, i.e., age, sex, diastolic blood pressure, smoking status, and self-reported physical exercise (Model 2). All models were set up with several independent vs. ΔaPWV as the dependent variable. Collinearity diagnostics was assessed in multiple regression models with no variable having a variance inflation factor above 3 (data not shown, available on request). IBM SPSS Statistics (version 22; IBM, Armonk, NY) was used for all statistical analyses. RESULTS Reduction in PWV following long-term weight loss Clinical characteristics of the participants at baseline and follow-up visits are presented in Table 1. At follow-up, average body weight and all examined anthropometric/metabolic parameters and WAT characteristics had significantly improved, as did aPWV. Table 1. Clinical characteristics of the study population before and 2 years after bariatric surgery   All individuals  Individuals for gene expression analysis    N  Before  After  N  Before  After    Before/after  N or mean (SD)  N or mean (SD)  Before/after  N or mean (SD)  N or mean (SD)  Pulse wave velocity, m/s  82/82  7.80 (1.50)  7.23 (1.41)**  30/30  8.03 (1.67)  7.23 (1.32)**  Pulse pressure, mm Hg  82/82  55.9 (10.5)  46.5 (7.2)***  30/30  54.6 (9.3)  46.9 (7.6)**  Central systolic blood pressure, mm Hg  82/82  129.2 (18.6)  116.5 (17.8)***  30/30  128.8 (18.5)  118.5 (19.2)**  Augmentation index, %  82/82  28.3 (15.4)  29.0 (15.2)  30/30  25.5 (15.4)  26.1 (14.7)  Age, years  82/82  40.8 (9.5)  42.9 (9.5)***  30/30  40.3 (9.16)  42.4 (1.67)***  Sex, female  82/82  71  71  28/28  28  28  Self-reported exercise  80/80  1.8 (0.6)  2.3 (0.5)***  29/29  1.8 (0.7)  2.3 (0.5)**  Smoking, yes  81/80  6  17**  30/30  2  5**  BMI  82/82  39.4 (3.4)  26.6 (3.5)***  30/30  38,9 (2.8)  26.7 (3.8)***  Resting heart rate, bpm  82/82  66.9 (10.3)  58.5 (7.8)***  30/30  70.3 (11.9)  59.4 (9.8)***  Systolic blood pressure, mm Hg  82/82  133.7 (15.8)  119.4 (13.1)***  30/30  133.6 (14.3)  120.33 (15.2)***  Diastolic blood pressure, mm Hg  82/82  77.7 (11.9)  72.6 (9.3)***  29/29  79.8 (14.6)  74.1 (9.9)*  Waist–hip ratio  81/81  0.97 (0.08)  0.89 (0.07)***  30/30  0.96 (0.07)  0.89 (0.07)***  Body fat, %  79/79  50.5 (4.6)  34.3 (7.4)***  30/30  50.8 (3.2)  34.9 (7.8)***  ESAT, g  72/72  3,411 (876)  1,406 (723)***  28/28  3,347 (665)  1,427 (778)***  EVAT, g  71/71  2,052 (877)  627 (504)***  28/28  1,974 (861)  646 (518)***  Visceral-to-subcutaneous fat mass ratio (EVAT/ESAT)  72/72  0.65 (0.42)  0.53 (0.39)*  28/28  0.62 (0.32)  0.55 (0.40)  Subcutaneous adipocyte volume, pl  73/73  873 (213)  339 (130)***  29/29  896 (223)  323 (131)***  Visceral adipocyte volume, pl  79/0  543 (203)  Not available  27  573 (216)  Not available  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  65/65  4.9 (1.3)  7.8 (1.8)***  28/28  5.3 (1.7)  8.2 (1.9)***  Fasting insulin levels, mIU  67/67  15.2 (9.5)  4.6 (1.3)***  24/24  13.8 (9.5)  4.4 (1.2)***  S-CRP, mg/l  74/74  5.7 (4.7)  0.6 (0.8)***  25/25  7.0 (7.0)  0.5 (0.5)***  S-Triglycerides, mmol/l  79/79  1.4 (.6)  0.8 (0.4)***  29/29  1.3 (0.5)  0.8 (0.3)***  S-FFA, mmol/l  67/67  0.7 (0.2)  0.6 (0.2)**  24/24  0.7 (0.2)  0.6 (0.2)*  ApoB/ApoA1  78/78  0.79 (0.24)  0.56 (0.16)***  27/27  0.82 (0.21)  0.58 (0.16)***  LDL cholesterol, mmol/l  79/79  2.9 (0.7)  2.1 (0.6)***  29/29  3.1 (0.6)  2.3 (0.8)***  P-fasting glucose, mmol/l  78/78  5.4 (0.9)  5.1 (0.4)***  29/29  5.4 (1.1)  5.1 (0.5)  Framingham risk score women  65/65  5.7 (5.8)  3.5 (6.4)***  27/27  5.3 (6.1)  3.3 (6.8)  Framingham risk score men  11/11  5.7 (5.8)  2.8 (8.1)**  2/2  9 (1)  5.5 (2.5)    All individuals  Individuals for gene expression analysis    N  Before  After  N  Before  After    Before/after  N or mean (SD)  N or mean (SD)  Before/after  N or mean (SD)  N or mean (SD)  Pulse wave velocity, m/s  82/82  7.80 (1.50)  7.23 (1.41)**  30/30  8.03 (1.67)  7.23 (1.32)**  Pulse pressure, mm Hg  82/82  55.9 (10.5)  46.5 (7.2)***  30/30  54.6 (9.3)  46.9 (7.6)**  Central systolic blood pressure, mm Hg  82/82  129.2 (18.6)  116.5 (17.8)***  30/30  128.8 (18.5)  118.5 (19.2)**  Augmentation index, %  82/82  28.3 (15.4)  29.0 (15.2)  30/30  25.5 (15.4)  26.1 (14.7)  Age, years  82/82  40.8 (9.5)  42.9 (9.5)***  30/30  40.3 (9.16)  42.4 (1.67)***  Sex, female  82/82  71  71  28/28  28  28  Self-reported exercise  80/80  1.8 (0.6)  2.3 (0.5)***  29/29  1.8 (0.7)  2.3 (0.5)**  Smoking, yes  81/80  6  17**  30/30  2  5**  BMI  82/82  39.4 (3.4)  26.6 (3.5)***  30/30  38,9 (2.8)  26.7 (3.8)***  Resting heart rate, bpm  82/82  66.9 (10.3)  58.5 (7.8)***  30/30  70.3 (11.9)  59.4 (9.8)***  Systolic blood pressure, mm Hg  82/82  133.7 (15.8)  119.4 (13.1)***  30/30  133.6 (14.3)  120.33 (15.2)***  Diastolic blood pressure, mm Hg  82/82  77.7 (11.9)  72.6 (9.3)***  29/29  79.8 (14.6)  74.1 (9.9)*  Waist–hip ratio  81/81  0.97 (0.08)  0.89 (0.07)***  30/30  0.96 (0.07)  0.89 (0.07)***  Body fat, %  79/79  50.5 (4.6)  34.3 (7.4)***  30/30  50.8 (3.2)  34.9 (7.8)***  ESAT, g  72/72  3,411 (876)  1,406 (723)***  28/28  3,347 (665)  1,427 (778)***  EVAT, g  71/71  2,052 (877)  627 (504)***  28/28  1,974 (861)  646 (518)***  Visceral-to-subcutaneous fat mass ratio (EVAT/ESAT)  72/72  0.65 (0.42)  0.53 (0.39)*  28/28  0.62 (0.32)  0.55 (0.40)  Subcutaneous adipocyte volume, pl  73/73  873 (213)  339 (130)***  29/29  896 (223)  323 (131)***  Visceral adipocyte volume, pl  79/0  543 (203)  Not available  27  573 (216)  Not available  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  65/65  4.9 (1.3)  7.8 (1.8)***  28/28  5.3 (1.7)  8.2 (1.9)***  Fasting insulin levels, mIU  67/67  15.2 (9.5)  4.6 (1.3)***  24/24  13.8 (9.5)  4.4 (1.2)***  S-CRP, mg/l  74/74  5.7 (4.7)  0.6 (0.8)***  25/25  7.0 (7.0)  0.5 (0.5)***  S-Triglycerides, mmol/l  79/79  1.4 (.6)  0.8 (0.4)***  29/29  1.3 (0.5)  0.8 (0.3)***  S-FFA, mmol/l  67/67  0.7 (0.2)  0.6 (0.2)**  24/24  0.7 (0.2)  0.6 (0.2)*  ApoB/ApoA1  78/78  0.79 (0.24)  0.56 (0.16)***  27/27  0.82 (0.21)  0.58 (0.16)***  LDL cholesterol, mmol/l  79/79  2.9 (0.7)  2.1 (0.6)***  29/29  3.1 (0.6)  2.3 (0.8)***  P-fasting glucose, mmol/l  78/78  5.4 (0.9)  5.1 (0.4)***  29/29  5.4 (1.1)  5.1 (0.5)  Framingham risk score women  65/65  5.7 (5.8)  3.5 (6.4)***  27/27  5.3 (6.1)  3.3 (6.8)  Framingham risk score men  11/11  5.7 (5.8)  2.8 (8.1)**  2/2  9 (1)  5.5 (2.5)  Data are shown for all individuals as well as for the subset of subjects where gene expression analysis was performed. Framingham risk score is expressed in points and are shown for men and women separately. A score <9 points for women indicates a 10 year risk for CVD below <1%. For men, a score of 9 points indicates a 10 year risk for CVD of 5%, a score of 5–6 indicates a 10 year risk for CVD of 2%, and a score of 1–4 indicates a 10 year risk for CVD of 1%. Abbreviations: ApoA1; Apolipoprotein A1; ApoB; apolipoprotein B; BMI, body mass index; bpm, beats per minute; CRP, C-reactive protein; CVD, cardiovascular disease; ESAT, estimated abdominal subcutaneous white adipose tissue; EVAT, estimated abdominal visceral white adipose tissue; FFA, free fatty acids; LDL, low-density lipoprotein; P, plasma; S, serum. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. View Large Baseline factors predicting improvements in aPWV following weight loss Of the studied baseline WAT and anthropometric/metabolic factors, BMI, EVAT, fasting insulin level, and subcutaneous adipocyte volume were significantly (positively) associated with ΔaPWV (Table 2), but after Bonferroni correction only subcutaneous adipocyte volume was significantly associated with ΔaPWV (P ≤ 0.0003). Baseline subcutaneous adipocyte volume remained independently associated with ΔaPWV after adjusting for confounders in both Model 1 (adjusted r2 = 0.211, P = 0.008) and Model 2 (adjusted r2 = 0.257, P = 0.014) (Table 3). Table 2. Simple linear regression of the association between baseline adipose and metabolic characteristics (set as independent variable) and changes in aortic pulse wave velocity (set as dependent variable) following weight loss   r value  P value  Bonferroni corrected P value  BMI  0.299  0.006  NS  Waist–hip ratio  0.044  0.693  NS  Body fat, %  0.011  0.919  NS  EVAT, g  0.244  0.036  NS  ESAT, g  0.179  0.125  NS  Visceral-to-subcutaneous fat mass ratio (EVAT/ ESAT)  0.117  0.316  NS  Subcutaneous adipocyte volume, pl  0.390  0.0003  0.0045  Visceral adipocyte volume, pl  0.109  0.341  NS  Fasting insulin levels, mIU  0.302  0.006  NS  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  0.012  0.921  NS  P-fasting glucose, mmol/l  0.113  0.315  NS  S-CRP, mg/l  0.012  0.919  NS  S-Triglycerides, mmol/l  0.047  0.677  NS  S-FFA, mmol/l  0.091  0.414  NS  LDL cholesterol, mmol/l  0.013  0.907  NS    r value  P value  Bonferroni corrected P value  BMI  0.299  0.006  NS  Waist–hip ratio  0.044  0.693  NS  Body fat, %  0.011  0.919  NS  EVAT, g  0.244  0.036  NS  ESAT, g  0.179  0.125  NS  Visceral-to-subcutaneous fat mass ratio (EVAT/ ESAT)  0.117  0.316  NS  Subcutaneous adipocyte volume, pl  0.390  0.0003  0.0045  Visceral adipocyte volume, pl  0.109  0.341  NS  Fasting insulin levels, mIU  0.302  0.006  NS  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  0.012  0.921  NS  P-fasting glucose, mmol/l  0.113  0.315  NS  S-CRP, mg/l  0.012  0.919  NS  S-Triglycerides, mmol/l  0.047  0.677  NS  S-FFA, mmol/l  0.091  0.414  NS  LDL cholesterol, mmol/l  0.013  0.907  NS  r and P values are shown, the latter also after Bonferroni correction (multiplied by 15). Abbreviations: BMI, body mass index; CRP, C-reactive protein; EVAT, estimated abdominal visceral white adipose tissue; ESAT, estimated abdominal subcutaneous white adipose tissue; FFA, free fatty acids; LDL, low-density lipoprotein; NS, not significant; P, plasma; S, serum. View Large Table 3. Multiple linear regression of the association between baseline subcutaneous adipocyte volume (set as independent variable) and change in aortic pulse wave velocity (set as dependent variable) following weight loss   Model 1  Model 2  Std beta  P value  Std beta  P value  Subcutaneous adipocyte volume  0.293  0.008  0.279  0.014    Model 1  Model 2  Std beta  P value  Std beta  P value  Subcutaneous adipocyte volume  0.293  0.008  0.279  0.014  Model 1 adjusted for systolic blood pressure and resting heart rate. Model 2 adjusted for systolic blood pressure, resting heart rate, diastolic blood pressure, age, sex, smoking status, and self-reported exercise. Standardized beta-coefficients (Std beta) and P values for subcutaneous volume are given for the 2 models. View Large Subcutaneous adipose tissue gene expression and aPWV improvement Analysis of global gene expression found no significant associations with ΔaPWV after correction for multiple testing. However, Bonferroni correction applied to the analysis of a preselected list of genes previously linked to PWV (detailed in Materials and Methods, Supplementary Table S1) showed that expression of COL4A1, was significantly associated with ΔaPWV (P = 0.0010). Baseline expression of COL4A1 remained independently associated with ΔaPWV after adjusting for confounders both in Model 1 (adjusted r2 = 0.504, P = 0.001) and Model 2 (adjusted r2 = 0.506, P = 0.006) (Table 4). Table 4. Multiple linear regression of the association between baseline COL4A1 gene expression (set as independent variable) and changes in pulse wave velocity (set as dependent variable) following weight loss   Model 1  Model 2  Std beta  P value  Std beta  P value  COL4A1 expression  0.534  0.001  0.486  0.006    Model 1  Model 2  Std beta  P value  Std beta  P value  COL4A1 expression  0.534  0.001  0.486  0.006  Model 1 adjusted for systolic blood pressure and resting heart rate. Model 2 adjusted for systolic blood pressure, resting heart rate, diastolic blood pressure, age, sex, smoking status, and self-reported exercise. Standardized beta-coefficients (Std beta) and P values for COL4A1 expression are given for the 2 models. View Large Improvements in adipose tissue factors associating with improvements in aPWV Our data indicated that adipose tissue phenotype predicted changes in aPWV. We also explored if changes in BMI, body composition/fat distribution, and subcutaneous adipocyte volume associated with changes in aPWV by simple linear regression (Supplementary Table S2). Unfortunately, follow-up data on gene expression were not available. Reductions in BMI, body fat % or waist-hip ratio did not associate with changes in aPWV. However, changes in subcutaneous adipocyte volume and EVAT were significantly associated with improvements in aPWV for all 82 subjects (subcutaneous adipocyte volume r = 0.318, P = 0.006; EVAT r = 0.331, P = 0.005), also after correction for multiple testing. Among the 30 individuals where gene expression was performed at baseline, only changes in subcutaneous adipocyte volume remained significant after Bonferroni correction (r = 0.518, P = 0.004). Effects of weight loss on other vascular measures After weight loss, subjects improved significantly in PP and SBPao but there was no change in AIX (Table 1). Simple linear regression was performed between baseline adipose tissue markers and change in PP and SBPao, respectively. Subcutaneous adipocyte volume associated with changes in PP (r = 0.268, P = 0.016) while baseline body fat % (r = 0.242, P = 0.030), ESAT (r = 0.306, P = 0.008), and EVAT/ESAT ratio (r = 0.240, P = 0.038), associated with changes in SBPao. After adjusting for resting heart rate, the relationships for body fat %, ESAT (r = 0.315, P = 0.010) and EVAT/ESAT ratio (r = 0.274, P = 0.030) were still significant. ESAT also remained significant after a further adjustment for baseline smoking status, self-reported exercise, age, and gender (r = 0.430, P = 0.009). The expression of COL4A1 did not associate with improvements in either PP or SBPao. DISCUSSION To our knowledge, this is the first long-term study of the effects of pronounced weight loss on arterial stiffness. Our main finding is that weight loss was associated with an improvement in aPWV by an average of 0.58 m/s. While this reduction may seem small, it could be influenced by the cohort’s relatively normal baseline aPWV (as compared to a healthy reference population where subjects 40–49 years of age had mean (±2 SD) values of 7.2 m/s (4.6–9.8).28 Nevertheless, the improvement was larger compared to the effects of aerobic exercise training, where a systematic review found the weighted mean difference to be −0.39 m/s (95% confidence interval: −0.52, −0.27).4 Admittedly, there are very few outcome studies on the effects of reducing PWV, we are only aware of one, showing that failure to lower PWV leads to increased mortality in end-stage renal disease patients.29 However, an increase of 1 m/s in aPWV increases both cardiovascular and all-cause mortality by 15%,1 suggesting that our findings are likely to be clinically significant. We also report that baseline subcutaneous adipocyte volume and COL4A1 expression in WAT were the only WAT-related characteristics that predicted the improvement in PWV after adjustment for multiple comparisons or established confounders. While our study design cannot prove causality, these results indicate that WAT may play a pathophysiological role in arterial stiffness. Previous cross-sectional studies have suggested that visceral WAT mass is associated with arterial stiffness.9 In the present longitudinal follow-up, baseline EVAT did not predict changes in aPWV, although there was an association between reduction in EVAT and improvements in aPWV. However, this was the only measure of body fat distribution or body composition that associated with changes in aPWV after correction for multiple comparisons. Moreover, a meta-analysis of weight reduction studies showed that changes in PWV did not associate with baseline weight or weight change.16 Altogether, this suggests that fat mass per se is not a major factor influencing PWV. Instead our findings, together with our previous data showing that adipocyte volume rather than fat mass is associated with aPWV,14 suggest that WAT morphology affects PWV. Several mechanisms could explain this. For example, a larger adipocyte volume has been linked to a more proinflammatory phenotype,30 which may contribute to arterial stiffening. In addition, fat tissue consisting of larger adipocytes have less capacity to store lipids, which leads to ectopic lipid deposition31 and may also contribute to arterial stiffness. It is known that subcutaneous adipocyte volume is associated with increased insulin resistance,32 and a correlation between insulin resistance and arterial stiffness has been suggested,7 However, this mechanism is less likely as we did not observe any associations between ΔaPWV and measures of insulin sensitivity. Further evidence supporting a link between WAT and aPWV was provided by the association between COL4A1 expression and ΔaPWV. After adjusting for established confounders, COL4A1 expression explained ~25% of the variations in aPWV improvement (i.e., Std beta-coeff2). COL4A1 encodes a collagen α1 chain, which, together with α2 chains, forms type IV collagen.33 The latter makes up the basement membrane that supports blood vessels as well as other tissue types. COL4A1 was first reported to be associated with arterial stiffness in a genome-wide association studies.26 Baseline expression of COL4A1 in lymphoblastoid cell lines correlated positively with a progressive increase in carotid–femoral PWV in a 4-year longitudinal cohort study of 121 female twins.25 Although the mechanism that explains the link between COL4A1 and arterial stiffening is unknown, some authors have speculated that degradation of type IV collagen by type 2 matrix metalloprotease, a collagenase secreted and activated by vascular smooth muscle cells, permits vascular smooth muscle cells to enter the subendothelial space where they produce more extracellular matrix.26 This may increase arterial stiffness by increasing intima thickness and by altering endothelial function.34 Others have reported that advanced glycation end products, which are also linked to arterial stiffness, interfere with type IV collagen self-assembly.35 Also, COL4A1 mRNA level correlates with the expression of transforming growth factor beta-1 and -3,36 profibrotic growth factors that are upregulated in the thickened intima of stiffened arteries.37 Taken together, while it is unclear whether WAT-expressed COL4A1 plays a causal role, it appears to constitute an indirect measure of arterial stiffness. Interestingly, we found no association between COL4A1 and subcutaneous adipocyte volume, which suggests that these 2 factors may affect PWV via different mechanisms. The Arteriograph provided additional vascular measures, with the cohort improving in both PP and SBPao, but not in AIX. Several baseline adipose characteristics were associated with these improvements further suggesting a role for WAT in arterial dysfunction. There is an ongoing debate regarding the value of different arterial stiffness measurements where PWV has been shown to have the best predictive value with regard to cardiovascular outcome.2 AIX, SBPao, PWV, and PP provide complementary information about the arterial circulation. In support of this notion, a previous study comparing invasively measured AIX, PWV, and aortic pressure, found no significant relationship between AIX and PWV.38 There are several limitations to our study. The method we used to measure aPWV is based on brachial artery oscillations, a method so far lacking cardiovascular outcome data.21 Nevertheless, the Arteriograph has been used in more than 70 original publications, several of which have demonstrated the reproducibility39 and validity in comparison to gold standard PWV measured either invasively40 or noninvasively (using Complior and/or SphygmoCor)39,41–44 suggesting that it provides a reliable measure of PWV in a research setting. A summary of studies comparing Arteriograph-measured arterial stiffness indices with other methods is described in the online supplement and detailed in Supplementary Table S3. Furthermore, the Arteriograph has also been used in several cross-sectional and prospective studies in the cardiovascular field.45–52 It was outside the scope of the present work to further compare the Arteriograph with other measures of PWV. Although noninvasive blood pressure recordings require correct cuff/bladder dimensions to give accurate results, PWV and AIX are not dependent on a specific relation between arm circumference and cuff/bladder dimensions. These measures are rather dependent on artifact-free pulse recordings where peaks can be defined properly by the software of the equipment. Therefore, PWV measurements obtained by brachial oscillations should not be influenced by body composition changes induced by weight reduction. This was further substantiated by our observation that there was no association between reductions in BMI, body fat % or waist-hip ratio, and improvements in aPWV. Furthermore, results obtained using this method have been shown to correlate with invasive and noninvasive measurements of aPWV39,40 and central blood pressure in normal weight and obese individuals.40,53,54 Thus, Arteriograph-based aPWV measures are likely to provide reliable results independent of body weight. Aortic PWV values are dependent on correct measurements of aortic length. We estimated this as a straight line from the suprapubic notch to the pubic bone as this is considered to provide the most accurate estimate of the aortic pulse wave travel length from the assumed point of reflection around the aortic bifurcation.40 This type of estimation may give a small overestimation of the aortic length as measured by magnetic resonance imaging.55 We avoided variations in aortic length measurement as part of the PWV result by using the initially estimated length also at the follow-up investigation. Although the aortic length may increase over time, such change should be minimal during the study period; moreover, it mainly takes place in the ascending aorta, which is not involved in the calculation of PWV.56 Thus, the alteration of PWV that we have shown can only be explained by an alteration of transit time. The primary outcome in this study was change in aPWV from a baseline value and our results may therefore be influenced by regression towards the mean. However, we believe that this effect is limited because study inclusion and exclusion criteria were not based on aPWV values and all aPWV measures were based on the mean of several individual measurements at each time point. Furthermore, the majority of the subjects improved in PWV over 2 years and the significant difference in aPWV for the entire cohort (P = 0.001) suggest that it is not a random effect. In addition to these considerations, our study included primarily women, which reflects the fact that only a minority of obese subjects undergoing bariatric surgery in Sweden are men. We did not include an obese control group who did not undergo surgery. However, increasing age is a well-known determinant of arterial stiffness, and it is unlikely that spontaneous improvements in PWV would be observed in weight-stable middle-aged individuals over a 2-year observation. Finally, transcriptomic data was only obtained from the subcutaneous WAT depot and we cannot exclude depot-specific differences. In conclusion, our study demonstrates that bariatric surgery improves long-term arterial stiffness. The improvement in aPWV was predicted by subcutaneous adipocyte size and adipose tissue COL4A1 mRNA expression. Although the pathophysiological role of these factors and their value in predicting cardiovascular morbidity/mortality need to be explored in future studies, our results suggest an important link between WAT phenotype and arterial stiffness. SUPPLEMENTARY MATERIAL Supplementary data are available at American Journal of Hypertension online. DISCLOSURE The authors declared no conflict of interest. ACKNOWLEDGMENTS We are grateful for technical assistance provided by Yvonne Widlund, Katarina Hertel, Britt-Marie Leijonhufvud, Elisabeth Dungner, Eva Sjölin, Kerstin Wåhlén, and Gaby Åström. This study was supported by grants from the Swedish Research Council, Novo Nordisk Foundation including the Tripartite Immuno-metabolism Consortium (TrIC) Grant Number NNF15CC0018486 and the MSAM consortium NNF15SA0018346, CIMED (project title: Adipose tissue turnover – mechanisms and clinical impact), Swedish Diabetes Foundation (grant number DIA2016-097), EFSD (project title: A translational approach to identify novel regulators of insulin sensitivity in human adipose tissue), Stockholm County Council (grant numbers 20150517 and K2014-54X-14510-12-5), the Diabetes Research Program at Karolinska Institutet (grant number 2009-1068), Swedish Heart Lung Foundation (grant number 20150423) and The Erling-Persson Family Foundation (grant number 140607). REFERENCES 1. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol  2010; 55: 1318– 1327. Google Scholar CrossRef Search ADS PubMed  2. Mitchell GF, Hwang SJ, Vasan RS, Larson MG, Pencina MJ, Hamburg NM, Vita JA, Levy D, Benjamin EJ. Arterial stiffness and cardiovascular events: the Framingham Heart Study. Circulation  2010; 121: 505– 511. Google Scholar CrossRef Search ADS PubMed  3. Cecelja M, Chowienczyk P. Dissociation of aortic pulse wave velocity with risk factors for cardiovascular disease other than hypertension: a systematic review. Hypertension  2009; 54: 1328– 1336. Google Scholar CrossRef Search ADS PubMed  4. Ashor AW, Lara J, Siervo M, Celis-Morales C, Mathers JC. Effects of exercise modalities on arterial stiffness and wave reflection: a systematic review and meta-analysis of randomized controlled trials. PLoS One  2014; 9: e110034. Google Scholar CrossRef Search ADS PubMed  5. Stefanadis C, Tsiamis E, Vlachopoulos C, Stratos C, Toutouzas K, Pitsavos C, Marakas S, Boudoulas H, Toutouzas P. Unfavorable effect of smoking on the elastic properties of the human aorta. Circulation  1997; 95: 31– 38. Google Scholar CrossRef Search ADS PubMed  6. Safar ME, Czernichow S, Blacher J. Obesity, arterial stiffness, and cardiovascular risk. J Am Soc Nephrol  2006; 17( 4 Suppl 2): S109– S111. Google Scholar CrossRef Search ADS PubMed  7. Webb DR, Khunti K, Silverman R, Gray LJ, Srinivasan B, Lacy PS, Williams B, Davies MJ. Impact of metabolic indices on central artery stiffness: independent association of insulin resistance and glucose with aortic pulse wave velocity. Diabetologia  2010; 53: 1190– 1198. Google Scholar CrossRef Search ADS PubMed  8. Sabbatini AR, Fontana V, Laurent S, Moreno H. An update on the role of adipokines in arterial stiffness and hypertension. J Hypertens  2015; 33: 435– 444. Google Scholar CrossRef Search ADS PubMed  9. Ferreira I, Snijder MB, Twisk JW, van Mechelen W, Kemper HC, Seidell JC, Stehouwer CD. Central fat mass versus peripheral fat and lean mass: opposite (adverse versus favorable) associations with arterial stiffness? The Amsterdam Growth and Health Longitudinal Study. J Clin Endocrinol Metab  2004; 89: 2632– 2639. Google Scholar CrossRef Search ADS PubMed  10. Farb MG, Bigornia S, Mott M, Tanriverdi K, Morin KM, Freedman JE, Joseph L, Hess DT, Apovian CM, Vita JA, Gokce N. Reduced adipose tissue inflammation represents an intermediate cardiometabolic phenotype in obesity. J Am Coll Cardiol  2011; 58: 232– 237. Google Scholar CrossRef Search ADS PubMed  11. Tabara Y, Takahashi Y, Setoh K, Kawaguchi T, Gotoh N, Terao C, Yamada R, Kosugi S, Sekine A, Nakayama T, Matsuda F. Synergistic association of elevated serum free fatty acid and glucose levels with large arterial stiffness in a general population: The Nagahama Study. Metabolism  2016; 65: 66– 72. Google Scholar CrossRef Search ADS PubMed  12. McEniery CM, Wilkinson IB, Johansen NB, Witte DR, Singh-Manoux A, Kivimaki M, Tabak AG, Brunner EJ, Shipley MJ. Nondiabetic glucometabolic status and progression of aortic stiffness: the Whitehall II Study. Diabetes Care  2017; 25: dc16– 1773. 13. Johansen NB, Vistisen D, Brunner EJ, Tabak AG, Shipley MJ, Wilkinson IB, McEniery CM, Roden M, Herder C, Kivimaki M, Witte DR. Determinants of aortic stiffness: 16-year follow-up of the Whitehall II study. PLoS One  2012; 7: 22. 14. Arner P, Bäckdahl J, Hemmingsson P, Stenvinkel P, Eriksson-Hogling D, Näslund E, Thorell A, Andersson DP, Caidahl K, Rydén M. Regional variations in the relationship between arterial stiffness and adipocyte volume or number in obese subjects. Int J Obes (Lond)  2015; 39: 222– 227. Google Scholar CrossRef Search ADS PubMed  15. Lacolley P, Challande P, Osborne-Pellegrin M, Regnault V. Genetics and pathophysiology of arterial stiffness. Cardiovasc Res  2009; 81: 637– 648. Google Scholar CrossRef Search ADS PubMed  16. Petersen KS, Blanch N, Keogh JB, Clifton PM. Effect of weight loss on pulse wave velocity: systematic review and meta-analysis. Arterioscler Thromb Vasc Biol  2015; 35: 243– 252. Google Scholar CrossRef Search ADS PubMed  17. Rider OJ, Tayal U, Francis JM, Ali MK, Robinson MR, Byrne JP, Clarke K, Neubauer S. The effect of obesity and weight loss on aortic pulse wave velocity as assessed by magnetic resonance imaging. Obesity  2010; 18: 2311– 2316. Google Scholar CrossRef Search ADS PubMed  18. Domienik-Karłowicz J, Lisik W, Rymarczyk Z, Dzikowska-Diduch O, Chmura A, Demkow U, Pruszczyk P. The short-term effect of bariatric surgery on non-invasive markers of artery function in patients with metabolic syndrome. Diabetol Metab Syndr  2015; 7: 76. Google Scholar CrossRef Search ADS PubMed  19. Gjevestad E, Hjelmesaeth J, Sandbu R, Nordstrand N. Effects of intensive lifestyle intervention and gastric bypass on aortic stiffness: a 1-year nonrandomized clinical study. Obesity  2015; 23: 37– 45. Google Scholar CrossRef Search ADS PubMed  20. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation  2002; 106: 3143– 3421. PubMed  21. Tomlinson LA. Methods for assessing arterial stiffness: technical considerations. Curr Opin Nephrol Hypertens  2012; 21: 655– 660. Google Scholar CrossRef Search ADS PubMed  22. Hirsch J, Gallian E. Methods for the determination of adipose cell size in man and animals. J Lipid Res  1968; 9: 110– 119. Google Scholar PubMed  23. Arner P, Andersson DP, Thorne A, Wiren M, Hoffstedt J, Naslund E, Thorell A, Ryden M. Variations in the size of the major omentum are primarily determined by fat cell number. J Clin Endocrinol Metab  2013; 98: 2012– 4106. Google Scholar CrossRef Search ADS   24. Rydén M, Hrydziuszko O, Mileti E, Raman A, Bornholdt J, Boyd M, Toft E, Qvist V, Näslund E, Thorell A, Andersson DP, Dahlman I, Gao H, Sandelin A, Daub CO, Arner P. The adipose transcriptional response to insulin is determined by obesity, not insulin sensitivity. Cell Rep  2016; 16: 2317– 2326. Google Scholar CrossRef Search ADS PubMed  25. Cecelja M, Jiang B, Mangino M, Spector TD, Chowienczyk PJ. Association of cross-sectional and longitudinal change in arterial stiffness with gene expression in the Twins UK Cohort. Hypertension  2016; 67: 70– 76. Google Scholar CrossRef Search ADS PubMed  26. Tarasov KV, Sanna S, Scuteri A, Strait JB, Orrù M, Parsa A, Lin PI, Maschio A, Lai S, Piras MG, Masala M, Tanaka T, Post W, O’Connell JR, Schlessinger D, Cao A, Nagaraja R, Mitchell BD, Abecasis GR, Shuldiner AR, Uda M, Lakatta EG, Najjar SS. COL4A1 is associated with arterial stiffness by genome-wide association scan. Circ Cardiovasc Genet  2009; 2: 151– 158. Google Scholar CrossRef Search ADS PubMed  27. Mitchell GF, Verwoert GC, Tarasov KV, Isaacs A, Smith AV, Yasmin, Rietzschel ER, Tanaka T, Liu Y, Parsa A, Najjar SS, O’Shaughnessy KM, Sigurdsson S, De Buyzere ML, Larson MG, Sie MP, Andrews JS, Post WS, Mattace-Raso FU, McEniery CM, Eiriksdottir G, Segers P, Vasan RS, van Rijn MJ, Howard TD, McArdle PF, Dehghan A, Jewell ES, Newhouse SJ, Bekaert S, Hamburg NM, Newman AB, Hofman A, Scuteri A, De Bacquer D, Ikram MA, Psaty BM, Fuchsberger C, Olden M, Wain LV, Elliott P, Smith NL, Felix JF, Erdmann J, Vita JA, Sutton-Tyrrell K, Sijbrands EJ, Sanna S, Launer LJ, De Meyer T, Johnson AD, Schut AF, Herrington DM, Rivadeneira F, Uda M, Wilkinson IB, Aspelund T, Gillebert TC, Van Bortel L, Benjamin EJ, Oostra BA, Ding J, Gibson Q, Uitterlinden AG, Abecasis GR, Cockcroft JR, Gudnason V, De Backer GG, Ferrucci L, Harris TB, Shuldiner AR, van Duijn CM, Levy D, Lakatta EG, Witteman JC. Common genetic variation in the 3’-BCL11B gene desert is associated with carotid-femoral pulse wave velocity and excess cardiovascular disease risk: the AortaGen Consortium. Circ Cardiovasc Genet  2012; 5: 81– 90. Google Scholar CrossRef Search ADS PubMed  28. Reference Values for Arterial Stiffness’ Collaboration, Mattace-Raso FHA, Verwoert GC, Wittemana JC, Wilkinson I, Cockcroft J, McEniery C, Yasmin, Laurent S, Boutouyrie P, Bozec E, Hansen TW, Torp-Pedersen C, Ibsen H, Jeppesen J, Vermeersch SJ, Rietzschel E, De Buyzere M, Gillebert TC, Van Bortel L, Segers P, Vlachopoulos C, Aznaouridis C, Stefanadis C, Benetos A, Labat C, Lacolley P, Stehouwer C, Nijpels G, Dekker JM, Stehouwer C, Ferreira I, Twisk JW, Czernichow S, Galan P, Hercberg S, Pannier B, Guérin A, London G, Cruickshank JK, Anderson SG, Paini A, Agabiti Rosei E, Muiesan ML, Salvetti M, Filipovsky J, Seidlerova J, Dolejsova M. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J  2010; 31: 2338– 2350. Google Scholar CrossRef Search ADS PubMed  29. Guerin AP, Blacher J, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness attenuation on survival of patients in end-stage renal failure. Circulation  2001; 103: 987– 992. Google Scholar CrossRef Search ADS PubMed  30. Arner E, Ryden M, Arner P. Tumor necrosis factor alpha and regulation of adipose tissue. N Engl J Med  2010; 362: 1151– 1153. Google Scholar CrossRef Search ADS PubMed  31. Virtue S, Vidal-Puig A. Adipose tissue expandability, lipotoxicity and the Metabolic Syndrome–an allostatic perspective. Biochim Biophys Acta  2010; 1801: 338– 349. Google Scholar CrossRef Search ADS PubMed  32. Hoffstedt J, Arner E, Wahrenberg H, Andersson DP, Qvisth V, Löfgren P, Rydén M, Thörne A, Wirén M, Palmér M, Thorell A, Toft E, Arner P. Regional impact of adipose tissue morphology on the metabolic profile in morbid obesity. Diabetologia  2010; 53: 2496– 2503. Google Scholar CrossRef Search ADS PubMed  33. Timpl R. Structure and biological activity of basement membrane proteins. Eur J Biochem  1989; 180: 487– 502. Google Scholar CrossRef Search ADS PubMed  34. Wang MLE. Central Arterial Aging: Humans to Molecules Hypertension in the Elderly , Vol. 23. Amsterdam, The Netherlands: Elsevier, 2006. 35. Tsilibary EC, Charonis AS, Reger LA, Wohlhueter RM, Furcht LT. The effect of nonenzymatic glucosylation on the binding of the main noncollagenous NC1 domain to type IV collagen. J Biol Chem  1988; 263: 4302– 4308. Google Scholar PubMed  36. Reggio S, Rouault C, Poitou C, Bichet JC, Prifti E, Bouillot JL, Rizkalla S, Lacasa D, Tordjman J, Clément K. Increased basement membrane components in adipose tissue during obesity: links with TGFβ and metabolic phenotypes. J Clin Endocrinol Metab  2016; 101: 2578– 2587. Google Scholar CrossRef Search ADS PubMed  37. Lakatta EG. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: Part III: cellular and molecular clues to heart and arterial aging. Circulation  2003; 107: 490– 497. Google Scholar CrossRef Search ADS PubMed  38. Sakurai M, Yamakado T, Kurachi H, Kato T, Kuroda K, Ishisu R, Okamoto S, Isaka N, Nakano T, Ito M. The relationship between aortic augmentation index and pulse wave velocity: an invasive study. J Hypertens  2007; 25: 391– 397. Google Scholar CrossRef Search ADS PubMed  39. Baulmann J, Schillings U, Rickert S, Uen S, Düsing R, Illyes M, Cziraki A, Nickering G, Mengden T. A new oscillometric method for assessment of arterial stiffness: comparison with tonometric and piezo-electronic methods. J Hypertens  2008; 26: 523– 528. Google Scholar CrossRef Search ADS PubMed  40. Horváth IG, Németh A, Lenkey Z, Alessandri N, Tufano F, Kis P, Gaszner B, Cziráki A. Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity. J Hypertens  2010; 28: 2068– 2075. Google Scholar CrossRef Search ADS PubMed  41. Ikonomidis I, Ntai K, Kadoglou NP, Papadakis I, Kornelakis M, Tritakis V, Varoudi M, Papadima T, Triantafyllidi H, Parissis J, Paraskevaidis I, Lekakis J. The evaluation of pulse wave velocity using Arteriograph and Complior apparatus across multiple cohorts of cardiovascular-related diseases. Int J Cardiol  2013; 168: 4890– 4892. Google Scholar CrossRef Search ADS PubMed  42. Rajzer MW, Wojciechowska W, Klocek M, Palka I, Brzozowska-Kiszka M, Kawecka-Jaszcz K. Comparison of aortic pulse wave velocity measured by three techniques: Complior, SphygmoCor and Arteriograph. J Hypertens  2008; 26: 2001– 2007. Google Scholar CrossRef Search ADS PubMed  43. Jatoi NA, Mahmud A, Bennett K, Feely J. Assessment of arterial stiffness in hypertension: comparison of oscillometric (Arteriograph), piezoelectronic (Complior) and tonometric (SphygmoCor) techniques. J Hypertens  2009; 27: 2186– 2191. Google Scholar CrossRef Search ADS PubMed  44. Ring M, Eriksson MJ, Zierath JR, Caidahl K. Arterial stiffness estimation in healthy subjects: a validation of oscillometric (Arteriograph) and tonometric (SphygmoCor) techniques. Hypertens Res  2014; 37: 999– 1007. Google Scholar CrossRef Search ADS PubMed  45. Cruickshank JK, Silva MJ, Molaodi OR, Enayat ZE, Cassidy A, Karamanos A, Read UM, Faconti L, Dall P, Stansfield B, Harding S. Ethnic differences in and childhood influences on early adult pulse wave velocity: the determinants of adolescent, now Young Adult, Social Wellbeing, and Health Longitudinal Study. Hypertension  2016; 67: 1133– 1141. Google Scholar CrossRef Search ADS PubMed  46. Rezai MR, Wallace AM, Sattar N, Finn JD, Wu FC, Cruickshank JK. Ethnic differences in aortic pulse wave velocity occur in the descending aorta and may be related to vitamin D. Hypertension  2011; 58: 247– 253. Google Scholar CrossRef Search ADS PubMed  47. Fodor L, Premuzic V, Ivkovic V, Perkov D, Laganovic M, Vrkic TZ, Dika Z, Zivko M, Jelakovic B. Arterial stiffness in atherosclerotic renovascular hypertension. J Hypertens  2014; 32: 2238– 2245. Google Scholar CrossRef Search ADS PubMed  48. Tarnoki AD, Tarnoki DL, Stazi MA, Medda E, Cotichini R, Nisticò L, Fagnani C, Lucatelli P, Boatta E, Zini C, Fanelli F, Baracchini C, Meneghetti G, Osztovits J, Jermendy G, Préda I, Kiss RG, Metneki J, Horvath T, Karlinger K, Racz A, Lannert A, Molnar AA, Littvay L, Garami Z, Berczi V, Schillaci G. Heritability of central blood pressure and arterial stiffness: a twin study. J Hypertens  2012; 30: 1564– 1571. Google Scholar CrossRef Search ADS PubMed  49. Tarnoki A, Tarnoki DL, Godor E, Littvay L, Horvath T, Jermendy A, Merkely B, Jermendy G, Maurovich-Horvat P. 2d.05: relationship of coronary atherosclerosis with arterial stiffness and central systolic blood pressure. J Hypertens  2015; 33: 74. 50. Faconti L, Silva MJ, Molaodi OR, Enayat ZE, Cassidy A, Karamanos A, Nanino E, Read UM, Dall P, Stansfield B, Harding S, Cruickshank KJ. Can arterial wave augmentation in young adults help account for variability of cardiovascular risk in different British ethnic groups? J Hypertens  2016; 34: 2220– 2226. Google Scholar CrossRef Search ADS PubMed  51. Osman MW, Leone F, Nath M, Khalil A, Webb DR, Robinson TG, Mousa HA. Diurnal variation and repeatability of arterial stiffness and cardiac output measurements in the third trimester of uncomplicated pregnancy. J Hypertens  2017; 35: 2436– 2442. Google Scholar CrossRef Search ADS PubMed  52. Tarnoki AD, Tarnoki DL, Bogl LH, Medda E, Fagnani C, Nisticò L, Stazi MA, Brescianini S, Lucatelli P, Boatta E, Zini C, Fanelli F, Baracchini C, Meneghetti G, Osztovits J, Jermendy G, Kiss RG, Preda I, Karlinger K, Lannert A, Molnar AA, Littvay L, Garami Z, Berczi V, Pucci G, Baffy G, Schillaci G, Pietiläinen KH. Association of body mass index with arterial stiffness and blood pressure components: a twin study. Atherosclerosis  2013; 229: 388– 395. Google Scholar CrossRef Search ADS PubMed  53. Rossen NB, Laugesen E, Peters CD, Ebbehøj E, Knudsen ST, Poulsen PL, Bøtker HE, Hansen KW. Invasive validation of arteriograph estimates of central blood pressure in patients with type 2 diabetes. Am J Hypertens  2014; 27: 674– 679. Google Scholar CrossRef Search ADS PubMed  54. Echeverri D, Pizano A, Cabrales J, Moreno K. Validation of central and peripheral non-invasive hemodynamic variables using an oscillometric method. High Blood Press Cardiovasc Prev  2017; 28: 017– 0238. 55. Rezai MR, Cowan BR, Sherratt N, Finn JD, Wu FC, Cruickshank JK. A magnetic resonance perspective of the pulse wave transit time by the Arteriograph device and potential for improving aortic length estimation for central pulse wave velocity. Blood Press Monit  2013; 18: 111– 118. Google Scholar CrossRef Search ADS PubMed  56. Sugawara J, Hayashi K, Yokoi T, Tanaka H. Age-associated elongation of the ascending aorta in adults. JACC Cardiovasc Imaging  2008; 1: 739– 748. Google Scholar CrossRef Search ADS PubMed  © American Journal of Hypertension, Ltd 2017. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Hypertension Oxford University Press

Long-Term Improvement in Aortic Pulse Wave Velocity After Weight Loss Can Be Predicted by White Adipose Tissue Factors

Loading next page...
 
/lp/ou_press/long-term-improvement-in-aortic-pulse-wave-velocity-after-weight-loss-pclwb01Bfk
Publisher
Oxford University Press
Copyright
© American Journal of Hypertension, Ltd 2017. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0895-7061
eISSN
1941-7225
D.O.I.
10.1093/ajh/hpx201
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Arterial stiffness, measured by pulse wave velocity (PWV), is linked to obesity, cardiovascular disease, and all-cause mortality. Short-term weight loss improves PWV, but the long-term effects are unknown. We investigated the effect of pronounced long-term weight loss on PWV and whether anthropometric/metabolic parameters and/or white adipose tissue (WAT) phenotype could predict this change in PWV. METHODS Eighty-two obese subjects were examined before and 2 years after Roux-en-Y gastric bypass. Analyses included anthropometrics, routine clinical chemistry, and hyperinsulinemic-euglycemic clamp. Arterial stiffness was measured as aortic PWV (aPWV) using the Arteriograph device. WAT mass and distribution were assessed by dual-X-ray absorptiometry. Baseline visceral and subcutaneous WAT samples were obtained to measure adipocyte cell size. Transcriptomic profiling of subcutaneous WAT was performed in a subset of subjects (n = 30). RESULTS At the 2-year follow-up, there were significant decreases in body mass index (39.4 ± 3.5 kg/m2 vs. 26.6 ± 3.4 kg/m2; P < 0.0001) and aPWV (7.8 ± 1.5 m/s vs. 7.2 ± 1.4 m/s; P = 0.006). Multiple regression analyses showed that baseline subcutaneous adipocyte volume was associated with a reduction in aPWV (P = 0.014), after adjusting for confounders. Expression analyses of 52 genes implicated in arterial stiffness showed that only one, COL4A1, independently predicted improvements in aPWV after adjusting for confounders (P = 0.006). CONCLUSIONS Bariatric surgery leads to long-term reduction in aPWV. This improvement can be independently predicted by subcutaneous adipocyte volume and WAT COL4A1 expression, which suggests that subcutaneous WAT has a role in regulating aPWV. CLINICAL TRIALS REGISTRATION Trial Number NCT01727245 (clinicaltrials.gov) adipocyte/metabolism, bariatric surgery, blood pressure, cell size, hypertension, humans, longitudinal studies, morbid/complications, obesity, vascular stiffness Arterial stiffness, a measure of reduced arterial compliance following changes in intraluminal pressure, can be measured noninvasively by pulse wave velocity (PWV). PWV is a strong predictor of future cardiovascular events and all-cause mortality,1 independent of other risk factors.2 Age, heart rate, and blood pressure explain much of the interindividual variations in arterial stiffness,3 although physical activity and smoking also contribute.4,5 Body weight gain increases arterial stiffness.6 Therefore, numerous studies have explored the relationship between obesity and arterial stiffness. Cross-sectional studies, have reported positive associations between arterial stiffness and insulin resistance,7 adipokine secretion from white adipose tissue (WAT),8 WAT distribution,9 inflammation,10 and circulating levels of free fatty acids.11 Moreover, observational studies have identified associations between carotid–femoral PWV and Hemoglobin A1c (HbA1c) levels,12 and between aortic PWV (aPWV) and central obesity, plasma adiponectin, and triglyceride levels.13 We reported that visceral adipocyte volume is significantly associated with aPWV in obese individuals.14 Altogether, these findings suggest a possible causal relationship between WAT dysfunction and arterial stiffness. Genome-wide and candidate gene-association studies have identified numerous genes whose expression associates with arterial stiffness.15 These encode proteins in the renin–angiotensin–aldosterone pathway, extracellular matrix components, metalloproteinases, proinflammatory factors, β-adrenergic and endothelin signaling receptors, endothelial cell apoptosis, and vascular wall immune response. Although most of these are expressed in peripheral tissues, including WAT, it remains unclear whether expression of any of these genes in WAT is associated with arterial stiffness. There is accumulating evidence that short-term moderate weight reduction, by means of energy-restricted diet or energy-restricted diet combined with exercise in subjects followed for up to a year, improves arterial stiffness, as measured by PWV.16 However, studies in obese subjects undergoing bariatric surgery, which causes greater weight loss than diet and exercise, have reported divergent results. For example, 2 studies showed reduced PWV,17,18 while another suggested no improvement.19 Importantly, participants were followed for only a moderate time (up to 1 year), with subjects likely not weight stable at time of examinations. To our knowledge, no studies have examined the long-term effects of weight reduction on PWV. Herein, we aimed to determine whether weight loss results in a long-term reduction in aPWV. We also aimed to explore whether baseline WAT phenotype, including anthropometric/metabolic parameters or the expression of selected genes, could predict improvements in PWV. METHODS Subjects One-hundred-and-twenty obese individuals, scheduled for bariatric surgery, were consecutively recruited in a longitudinal study of the effects of weight loss on WAT function (clinicaltrials.gov, NCT01727245). The inclusion/exclusion criteria have been described elsewhere.14 Subjects underwent bariatric surgery with Roux-en-Y gastric bypass. Given that blood pressure has a major effect on arterial stiffness, subjects using antihypertensive medication (n = 37) were excluded. One patient was excluded because of a faulty aPWV measurement. Finally, 82 subjects (71 women and 11 men) were included. The study was approved by the regional ethics board in Stockholm and was conducted in accordance with the statutes of the Declaration of Helsinki. Written informed consent was obtained from all participants. Subjects were examined before and 24 months after surgery, when they had attained a stable body weight. Subjects came to the laboratory at 7.30 a.m. after an overnight fast and having abstained from nicotine and caffeine. Data on nicotine use, medications, self-assessed physical exercise (graded in predetermined increments on a scale from 1–4), and past medical history were collected in questionnaires and confirmed by interview. Standard anthropometric measures, including body weight, height, body mass index (BMI), and waist and hip circumferences, were recorded. Framingham risk scores were calculated according to Adult Treatment Panel III criteria.20 The same staff performed all measurements and laboratory examinations both before and 2 years after surgery. Arterial stiffness and blood pressure Recommended standardized procedures for measuring arterial stiffness were followed21 and were performed exactly as described.14 Resting heart rate was recorded after 15 minutes of rest in the supine position. Aortic PWV and blood pressure was measured using an Arteriograph (TensioMed, Budapest, Hungary) with appropriate cuff size based on subjects arm circumference at the time of examination. It was calculated by dividing the distance traveled by the pulse wave by its transit time (RT/2). Mean values based on at least 3 measurements were used. The distance traveled from the suprasternal notch to the pubic bone was measured in a straight line to avoid the influence of variations in abdominal circumference. As there is no evidence that aortic length is influenced by weight loss, the measure obtained at baseline was used for both baseline and 2-year aPWV calculations. The Arteriograph also provides other vascular measures; augmentation index (the % of central pulse pressure attributed to the reflected wave from the periphery) (AIX), central systolic blood pressure (the systolic blood pressure in the aorta) (SBPao), and pulse pressure (the difference between systolic and diastolic blood pressure) (PP). Systolic and diastolic blood pressures were also obtained using an automatic device (Omron M10-IT; Omron Health Care, Hoofddorp, The Netherlands). Assessment of in vivo insulin sensitivity and body composition Subjects underwent a hyperinsulinemic-euglycemic clamp and Dual-energy X-ray absorptiometry (DXA) exactly as previously described.14 All subjects were examined using a GE Lunar iDXA running the enCORE software (version 14.10.022) with the CoreScan feature (GE Medical Systems, Chalfont St. Giles, UK). This allowed determination of body fat composition (android, gynoid, estimated visceral adipose tissue; EVAT), and estimated subcutaneous adipose tissue (ESAT) in the android region as discussed previously.14 Of note, EVAT and ESAT do not represent total abdominal visceral and subcutaneous WAT, but only the region that corresponds to where the fat biopsy is obtained. Adipocyte cell volume Adipose tissue samples were treated with collagenase to isolate adipocytes. The diameter of 100 adipocytes was determined microscopically, and the mean adipocyte cell volume was calculated as described by Hirsch and Gallian.22 This method has been used in our laboratory for over 4 decades, and the validity has been described in detail elsewhere.23 Global transcriptomics of WAT Global transcriptomic profiling was performed using 5′cap analysis of gene expression (CAGE) of subcutaneous WAT from a subset of 30 participants at the baseline, as described previously in detail.24 In brief, total RNA was extracted and the quality was determined using the bioanalyzer RNA 6000 Pico Kit (Agilent Technologies). Messenger RNA levels are expressed as tag per million normalized read counts. All raw data are uploaded at https://export.uppmax.uu.se/b2013047/CellReportsTables/. Gene expression data were analyzed in a 2-step approach. First, a global analysis accepting a false discovery rate of 0.1 was performed. Because no significant associations were found, further analysis was undertaken by examining a previously used list of 52 genes found to be associated with arterial stiffness in genetic variant studies,25 which included candidate gene studies and genome-wide association studies.15,26,27 Statistics The data are expressed as mean ± SD. The Shapiro–Wilks test was used to identify normal distributions. Normally distributed values were compared using a paired-sample Student t test (2-tailed), and non-normally distributed values were compared using the related-samples Wilcoxon signed–rank test. Any missing data were excluded on a pairwise basis. Associations between baseline or changes in (i.e., delta; Δ) WAT phenotype/metabolic parameters with ΔaPWV were tested by simple linear regression analysis, where ΔaPWV was set as the dependent variable. Bonferroni correction was applied to correct for multiple testing resulting in alpha levels of 0.05/15 = 0.0033 (baseline WAT and metabolic factors vs. ΔaPWV) and 0.05/40 = 0.00125 (WAT gene expression and ΔaPWV). All parameters that were significantly associated in the simple regression were entered into a multiple regression analysis model adjusted for systolic blood pressure and resting heart rate (Model 1). In an additional analysis, we expanded Model 1 and included other factors known to associate with PWV, i.e., age, sex, diastolic blood pressure, smoking status, and self-reported physical exercise (Model 2). All models were set up with several independent vs. ΔaPWV as the dependent variable. Collinearity diagnostics was assessed in multiple regression models with no variable having a variance inflation factor above 3 (data not shown, available on request). IBM SPSS Statistics (version 22; IBM, Armonk, NY) was used for all statistical analyses. RESULTS Reduction in PWV following long-term weight loss Clinical characteristics of the participants at baseline and follow-up visits are presented in Table 1. At follow-up, average body weight and all examined anthropometric/metabolic parameters and WAT characteristics had significantly improved, as did aPWV. Table 1. Clinical characteristics of the study population before and 2 years after bariatric surgery   All individuals  Individuals for gene expression analysis    N  Before  After  N  Before  After    Before/after  N or mean (SD)  N or mean (SD)  Before/after  N or mean (SD)  N or mean (SD)  Pulse wave velocity, m/s  82/82  7.80 (1.50)  7.23 (1.41)**  30/30  8.03 (1.67)  7.23 (1.32)**  Pulse pressure, mm Hg  82/82  55.9 (10.5)  46.5 (7.2)***  30/30  54.6 (9.3)  46.9 (7.6)**  Central systolic blood pressure, mm Hg  82/82  129.2 (18.6)  116.5 (17.8)***  30/30  128.8 (18.5)  118.5 (19.2)**  Augmentation index, %  82/82  28.3 (15.4)  29.0 (15.2)  30/30  25.5 (15.4)  26.1 (14.7)  Age, years  82/82  40.8 (9.5)  42.9 (9.5)***  30/30  40.3 (9.16)  42.4 (1.67)***  Sex, female  82/82  71  71  28/28  28  28  Self-reported exercise  80/80  1.8 (0.6)  2.3 (0.5)***  29/29  1.8 (0.7)  2.3 (0.5)**  Smoking, yes  81/80  6  17**  30/30  2  5**  BMI  82/82  39.4 (3.4)  26.6 (3.5)***  30/30  38,9 (2.8)  26.7 (3.8)***  Resting heart rate, bpm  82/82  66.9 (10.3)  58.5 (7.8)***  30/30  70.3 (11.9)  59.4 (9.8)***  Systolic blood pressure, mm Hg  82/82  133.7 (15.8)  119.4 (13.1)***  30/30  133.6 (14.3)  120.33 (15.2)***  Diastolic blood pressure, mm Hg  82/82  77.7 (11.9)  72.6 (9.3)***  29/29  79.8 (14.6)  74.1 (9.9)*  Waist–hip ratio  81/81  0.97 (0.08)  0.89 (0.07)***  30/30  0.96 (0.07)  0.89 (0.07)***  Body fat, %  79/79  50.5 (4.6)  34.3 (7.4)***  30/30  50.8 (3.2)  34.9 (7.8)***  ESAT, g  72/72  3,411 (876)  1,406 (723)***  28/28  3,347 (665)  1,427 (778)***  EVAT, g  71/71  2,052 (877)  627 (504)***  28/28  1,974 (861)  646 (518)***  Visceral-to-subcutaneous fat mass ratio (EVAT/ESAT)  72/72  0.65 (0.42)  0.53 (0.39)*  28/28  0.62 (0.32)  0.55 (0.40)  Subcutaneous adipocyte volume, pl  73/73  873 (213)  339 (130)***  29/29  896 (223)  323 (131)***  Visceral adipocyte volume, pl  79/0  543 (203)  Not available  27  573 (216)  Not available  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  65/65  4.9 (1.3)  7.8 (1.8)***  28/28  5.3 (1.7)  8.2 (1.9)***  Fasting insulin levels, mIU  67/67  15.2 (9.5)  4.6 (1.3)***  24/24  13.8 (9.5)  4.4 (1.2)***  S-CRP, mg/l  74/74  5.7 (4.7)  0.6 (0.8)***  25/25  7.0 (7.0)  0.5 (0.5)***  S-Triglycerides, mmol/l  79/79  1.4 (.6)  0.8 (0.4)***  29/29  1.3 (0.5)  0.8 (0.3)***  S-FFA, mmol/l  67/67  0.7 (0.2)  0.6 (0.2)**  24/24  0.7 (0.2)  0.6 (0.2)*  ApoB/ApoA1  78/78  0.79 (0.24)  0.56 (0.16)***  27/27  0.82 (0.21)  0.58 (0.16)***  LDL cholesterol, mmol/l  79/79  2.9 (0.7)  2.1 (0.6)***  29/29  3.1 (0.6)  2.3 (0.8)***  P-fasting glucose, mmol/l  78/78  5.4 (0.9)  5.1 (0.4)***  29/29  5.4 (1.1)  5.1 (0.5)  Framingham risk score women  65/65  5.7 (5.8)  3.5 (6.4)***  27/27  5.3 (6.1)  3.3 (6.8)  Framingham risk score men  11/11  5.7 (5.8)  2.8 (8.1)**  2/2  9 (1)  5.5 (2.5)    All individuals  Individuals for gene expression analysis    N  Before  After  N  Before  After    Before/after  N or mean (SD)  N or mean (SD)  Before/after  N or mean (SD)  N or mean (SD)  Pulse wave velocity, m/s  82/82  7.80 (1.50)  7.23 (1.41)**  30/30  8.03 (1.67)  7.23 (1.32)**  Pulse pressure, mm Hg  82/82  55.9 (10.5)  46.5 (7.2)***  30/30  54.6 (9.3)  46.9 (7.6)**  Central systolic blood pressure, mm Hg  82/82  129.2 (18.6)  116.5 (17.8)***  30/30  128.8 (18.5)  118.5 (19.2)**  Augmentation index, %  82/82  28.3 (15.4)  29.0 (15.2)  30/30  25.5 (15.4)  26.1 (14.7)  Age, years  82/82  40.8 (9.5)  42.9 (9.5)***  30/30  40.3 (9.16)  42.4 (1.67)***  Sex, female  82/82  71  71  28/28  28  28  Self-reported exercise  80/80  1.8 (0.6)  2.3 (0.5)***  29/29  1.8 (0.7)  2.3 (0.5)**  Smoking, yes  81/80  6  17**  30/30  2  5**  BMI  82/82  39.4 (3.4)  26.6 (3.5)***  30/30  38,9 (2.8)  26.7 (3.8)***  Resting heart rate, bpm  82/82  66.9 (10.3)  58.5 (7.8)***  30/30  70.3 (11.9)  59.4 (9.8)***  Systolic blood pressure, mm Hg  82/82  133.7 (15.8)  119.4 (13.1)***  30/30  133.6 (14.3)  120.33 (15.2)***  Diastolic blood pressure, mm Hg  82/82  77.7 (11.9)  72.6 (9.3)***  29/29  79.8 (14.6)  74.1 (9.9)*  Waist–hip ratio  81/81  0.97 (0.08)  0.89 (0.07)***  30/30  0.96 (0.07)  0.89 (0.07)***  Body fat, %  79/79  50.5 (4.6)  34.3 (7.4)***  30/30  50.8 (3.2)  34.9 (7.8)***  ESAT, g  72/72  3,411 (876)  1,406 (723)***  28/28  3,347 (665)  1,427 (778)***  EVAT, g  71/71  2,052 (877)  627 (504)***  28/28  1,974 (861)  646 (518)***  Visceral-to-subcutaneous fat mass ratio (EVAT/ESAT)  72/72  0.65 (0.42)  0.53 (0.39)*  28/28  0.62 (0.32)  0.55 (0.40)  Subcutaneous adipocyte volume, pl  73/73  873 (213)  339 (130)***  29/29  896 (223)  323 (131)***  Visceral adipocyte volume, pl  79/0  543 (203)  Not available  27  573 (216)  Not available  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  65/65  4.9 (1.3)  7.8 (1.8)***  28/28  5.3 (1.7)  8.2 (1.9)***  Fasting insulin levels, mIU  67/67  15.2 (9.5)  4.6 (1.3)***  24/24  13.8 (9.5)  4.4 (1.2)***  S-CRP, mg/l  74/74  5.7 (4.7)  0.6 (0.8)***  25/25  7.0 (7.0)  0.5 (0.5)***  S-Triglycerides, mmol/l  79/79  1.4 (.6)  0.8 (0.4)***  29/29  1.3 (0.5)  0.8 (0.3)***  S-FFA, mmol/l  67/67  0.7 (0.2)  0.6 (0.2)**  24/24  0.7 (0.2)  0.6 (0.2)*  ApoB/ApoA1  78/78  0.79 (0.24)  0.56 (0.16)***  27/27  0.82 (0.21)  0.58 (0.16)***  LDL cholesterol, mmol/l  79/79  2.9 (0.7)  2.1 (0.6)***  29/29  3.1 (0.6)  2.3 (0.8)***  P-fasting glucose, mmol/l  78/78  5.4 (0.9)  5.1 (0.4)***  29/29  5.4 (1.1)  5.1 (0.5)  Framingham risk score women  65/65  5.7 (5.8)  3.5 (6.4)***  27/27  5.3 (6.1)  3.3 (6.8)  Framingham risk score men  11/11  5.7 (5.8)  2.8 (8.1)**  2/2  9 (1)  5.5 (2.5)  Data are shown for all individuals as well as for the subset of subjects where gene expression analysis was performed. Framingham risk score is expressed in points and are shown for men and women separately. A score <9 points for women indicates a 10 year risk for CVD below <1%. For men, a score of 9 points indicates a 10 year risk for CVD of 5%, a score of 5–6 indicates a 10 year risk for CVD of 2%, and a score of 1–4 indicates a 10 year risk for CVD of 1%. Abbreviations: ApoA1; Apolipoprotein A1; ApoB; apolipoprotein B; BMI, body mass index; bpm, beats per minute; CRP, C-reactive protein; CVD, cardiovascular disease; ESAT, estimated abdominal subcutaneous white adipose tissue; EVAT, estimated abdominal visceral white adipose tissue; FFA, free fatty acids; LDL, low-density lipoprotein; P, plasma; S, serum. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. View Large Baseline factors predicting improvements in aPWV following weight loss Of the studied baseline WAT and anthropometric/metabolic factors, BMI, EVAT, fasting insulin level, and subcutaneous adipocyte volume were significantly (positively) associated with ΔaPWV (Table 2), but after Bonferroni correction only subcutaneous adipocyte volume was significantly associated with ΔaPWV (P ≤ 0.0003). Baseline subcutaneous adipocyte volume remained independently associated with ΔaPWV after adjusting for confounders in both Model 1 (adjusted r2 = 0.211, P = 0.008) and Model 2 (adjusted r2 = 0.257, P = 0.014) (Table 3). Table 2. Simple linear regression of the association between baseline adipose and metabolic characteristics (set as independent variable) and changes in aortic pulse wave velocity (set as dependent variable) following weight loss   r value  P value  Bonferroni corrected P value  BMI  0.299  0.006  NS  Waist–hip ratio  0.044  0.693  NS  Body fat, %  0.011  0.919  NS  EVAT, g  0.244  0.036  NS  ESAT, g  0.179  0.125  NS  Visceral-to-subcutaneous fat mass ratio (EVAT/ ESAT)  0.117  0.316  NS  Subcutaneous adipocyte volume, pl  0.390  0.0003  0.0045  Visceral adipocyte volume, pl  0.109  0.341  NS  Fasting insulin levels, mIU  0.302  0.006  NS  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  0.012  0.921  NS  P-fasting glucose, mmol/l  0.113  0.315  NS  S-CRP, mg/l  0.012  0.919  NS  S-Triglycerides, mmol/l  0.047  0.677  NS  S-FFA, mmol/l  0.091  0.414  NS  LDL cholesterol, mmol/l  0.013  0.907  NS    r value  P value  Bonferroni corrected P value  BMI  0.299  0.006  NS  Waist–hip ratio  0.044  0.693  NS  Body fat, %  0.011  0.919  NS  EVAT, g  0.244  0.036  NS  ESAT, g  0.179  0.125  NS  Visceral-to-subcutaneous fat mass ratio (EVAT/ ESAT)  0.117  0.316  NS  Subcutaneous adipocyte volume, pl  0.390  0.0003  0.0045  Visceral adipocyte volume, pl  0.109  0.341  NS  Fasting insulin levels, mIU  0.302  0.006  NS  Insulin sensitivity, (glucose infusion rate) mg/kg min−1  0.012  0.921  NS  P-fasting glucose, mmol/l  0.113  0.315  NS  S-CRP, mg/l  0.012  0.919  NS  S-Triglycerides, mmol/l  0.047  0.677  NS  S-FFA, mmol/l  0.091  0.414  NS  LDL cholesterol, mmol/l  0.013  0.907  NS  r and P values are shown, the latter also after Bonferroni correction (multiplied by 15). Abbreviations: BMI, body mass index; CRP, C-reactive protein; EVAT, estimated abdominal visceral white adipose tissue; ESAT, estimated abdominal subcutaneous white adipose tissue; FFA, free fatty acids; LDL, low-density lipoprotein; NS, not significant; P, plasma; S, serum. View Large Table 3. Multiple linear regression of the association between baseline subcutaneous adipocyte volume (set as independent variable) and change in aortic pulse wave velocity (set as dependent variable) following weight loss   Model 1  Model 2  Std beta  P value  Std beta  P value  Subcutaneous adipocyte volume  0.293  0.008  0.279  0.014    Model 1  Model 2  Std beta  P value  Std beta  P value  Subcutaneous adipocyte volume  0.293  0.008  0.279  0.014  Model 1 adjusted for systolic blood pressure and resting heart rate. Model 2 adjusted for systolic blood pressure, resting heart rate, diastolic blood pressure, age, sex, smoking status, and self-reported exercise. Standardized beta-coefficients (Std beta) and P values for subcutaneous volume are given for the 2 models. View Large Subcutaneous adipose tissue gene expression and aPWV improvement Analysis of global gene expression found no significant associations with ΔaPWV after correction for multiple testing. However, Bonferroni correction applied to the analysis of a preselected list of genes previously linked to PWV (detailed in Materials and Methods, Supplementary Table S1) showed that expression of COL4A1, was significantly associated with ΔaPWV (P = 0.0010). Baseline expression of COL4A1 remained independently associated with ΔaPWV after adjusting for confounders both in Model 1 (adjusted r2 = 0.504, P = 0.001) and Model 2 (adjusted r2 = 0.506, P = 0.006) (Table 4). Table 4. Multiple linear regression of the association between baseline COL4A1 gene expression (set as independent variable) and changes in pulse wave velocity (set as dependent variable) following weight loss   Model 1  Model 2  Std beta  P value  Std beta  P value  COL4A1 expression  0.534  0.001  0.486  0.006    Model 1  Model 2  Std beta  P value  Std beta  P value  COL4A1 expression  0.534  0.001  0.486  0.006  Model 1 adjusted for systolic blood pressure and resting heart rate. Model 2 adjusted for systolic blood pressure, resting heart rate, diastolic blood pressure, age, sex, smoking status, and self-reported exercise. Standardized beta-coefficients (Std beta) and P values for COL4A1 expression are given for the 2 models. View Large Improvements in adipose tissue factors associating with improvements in aPWV Our data indicated that adipose tissue phenotype predicted changes in aPWV. We also explored if changes in BMI, body composition/fat distribution, and subcutaneous adipocyte volume associated with changes in aPWV by simple linear regression (Supplementary Table S2). Unfortunately, follow-up data on gene expression were not available. Reductions in BMI, body fat % or waist-hip ratio did not associate with changes in aPWV. However, changes in subcutaneous adipocyte volume and EVAT were significantly associated with improvements in aPWV for all 82 subjects (subcutaneous adipocyte volume r = 0.318, P = 0.006; EVAT r = 0.331, P = 0.005), also after correction for multiple testing. Among the 30 individuals where gene expression was performed at baseline, only changes in subcutaneous adipocyte volume remained significant after Bonferroni correction (r = 0.518, P = 0.004). Effects of weight loss on other vascular measures After weight loss, subjects improved significantly in PP and SBPao but there was no change in AIX (Table 1). Simple linear regression was performed between baseline adipose tissue markers and change in PP and SBPao, respectively. Subcutaneous adipocyte volume associated with changes in PP (r = 0.268, P = 0.016) while baseline body fat % (r = 0.242, P = 0.030), ESAT (r = 0.306, P = 0.008), and EVAT/ESAT ratio (r = 0.240, P = 0.038), associated with changes in SBPao. After adjusting for resting heart rate, the relationships for body fat %, ESAT (r = 0.315, P = 0.010) and EVAT/ESAT ratio (r = 0.274, P = 0.030) were still significant. ESAT also remained significant after a further adjustment for baseline smoking status, self-reported exercise, age, and gender (r = 0.430, P = 0.009). The expression of COL4A1 did not associate with improvements in either PP or SBPao. DISCUSSION To our knowledge, this is the first long-term study of the effects of pronounced weight loss on arterial stiffness. Our main finding is that weight loss was associated with an improvement in aPWV by an average of 0.58 m/s. While this reduction may seem small, it could be influenced by the cohort’s relatively normal baseline aPWV (as compared to a healthy reference population where subjects 40–49 years of age had mean (±2 SD) values of 7.2 m/s (4.6–9.8).28 Nevertheless, the improvement was larger compared to the effects of aerobic exercise training, where a systematic review found the weighted mean difference to be −0.39 m/s (95% confidence interval: −0.52, −0.27).4 Admittedly, there are very few outcome studies on the effects of reducing PWV, we are only aware of one, showing that failure to lower PWV leads to increased mortality in end-stage renal disease patients.29 However, an increase of 1 m/s in aPWV increases both cardiovascular and all-cause mortality by 15%,1 suggesting that our findings are likely to be clinically significant. We also report that baseline subcutaneous adipocyte volume and COL4A1 expression in WAT were the only WAT-related characteristics that predicted the improvement in PWV after adjustment for multiple comparisons or established confounders. While our study design cannot prove causality, these results indicate that WAT may play a pathophysiological role in arterial stiffness. Previous cross-sectional studies have suggested that visceral WAT mass is associated with arterial stiffness.9 In the present longitudinal follow-up, baseline EVAT did not predict changes in aPWV, although there was an association between reduction in EVAT and improvements in aPWV. However, this was the only measure of body fat distribution or body composition that associated with changes in aPWV after correction for multiple comparisons. Moreover, a meta-analysis of weight reduction studies showed that changes in PWV did not associate with baseline weight or weight change.16 Altogether, this suggests that fat mass per se is not a major factor influencing PWV. Instead our findings, together with our previous data showing that adipocyte volume rather than fat mass is associated with aPWV,14 suggest that WAT morphology affects PWV. Several mechanisms could explain this. For example, a larger adipocyte volume has been linked to a more proinflammatory phenotype,30 which may contribute to arterial stiffening. In addition, fat tissue consisting of larger adipocytes have less capacity to store lipids, which leads to ectopic lipid deposition31 and may also contribute to arterial stiffness. It is known that subcutaneous adipocyte volume is associated with increased insulin resistance,32 and a correlation between insulin resistance and arterial stiffness has been suggested,7 However, this mechanism is less likely as we did not observe any associations between ΔaPWV and measures of insulin sensitivity. Further evidence supporting a link between WAT and aPWV was provided by the association between COL4A1 expression and ΔaPWV. After adjusting for established confounders, COL4A1 expression explained ~25% of the variations in aPWV improvement (i.e., Std beta-coeff2). COL4A1 encodes a collagen α1 chain, which, together with α2 chains, forms type IV collagen.33 The latter makes up the basement membrane that supports blood vessels as well as other tissue types. COL4A1 was first reported to be associated with arterial stiffness in a genome-wide association studies.26 Baseline expression of COL4A1 in lymphoblastoid cell lines correlated positively with a progressive increase in carotid–femoral PWV in a 4-year longitudinal cohort study of 121 female twins.25 Although the mechanism that explains the link between COL4A1 and arterial stiffening is unknown, some authors have speculated that degradation of type IV collagen by type 2 matrix metalloprotease, a collagenase secreted and activated by vascular smooth muscle cells, permits vascular smooth muscle cells to enter the subendothelial space where they produce more extracellular matrix.26 This may increase arterial stiffness by increasing intima thickness and by altering endothelial function.34 Others have reported that advanced glycation end products, which are also linked to arterial stiffness, interfere with type IV collagen self-assembly.35 Also, COL4A1 mRNA level correlates with the expression of transforming growth factor beta-1 and -3,36 profibrotic growth factors that are upregulated in the thickened intima of stiffened arteries.37 Taken together, while it is unclear whether WAT-expressed COL4A1 plays a causal role, it appears to constitute an indirect measure of arterial stiffness. Interestingly, we found no association between COL4A1 and subcutaneous adipocyte volume, which suggests that these 2 factors may affect PWV via different mechanisms. The Arteriograph provided additional vascular measures, with the cohort improving in both PP and SBPao, but not in AIX. Several baseline adipose characteristics were associated with these improvements further suggesting a role for WAT in arterial dysfunction. There is an ongoing debate regarding the value of different arterial stiffness measurements where PWV has been shown to have the best predictive value with regard to cardiovascular outcome.2 AIX, SBPao, PWV, and PP provide complementary information about the arterial circulation. In support of this notion, a previous study comparing invasively measured AIX, PWV, and aortic pressure, found no significant relationship between AIX and PWV.38 There are several limitations to our study. The method we used to measure aPWV is based on brachial artery oscillations, a method so far lacking cardiovascular outcome data.21 Nevertheless, the Arteriograph has been used in more than 70 original publications, several of which have demonstrated the reproducibility39 and validity in comparison to gold standard PWV measured either invasively40 or noninvasively (using Complior and/or SphygmoCor)39,41–44 suggesting that it provides a reliable measure of PWV in a research setting. A summary of studies comparing Arteriograph-measured arterial stiffness indices with other methods is described in the online supplement and detailed in Supplementary Table S3. Furthermore, the Arteriograph has also been used in several cross-sectional and prospective studies in the cardiovascular field.45–52 It was outside the scope of the present work to further compare the Arteriograph with other measures of PWV. Although noninvasive blood pressure recordings require correct cuff/bladder dimensions to give accurate results, PWV and AIX are not dependent on a specific relation between arm circumference and cuff/bladder dimensions. These measures are rather dependent on artifact-free pulse recordings where peaks can be defined properly by the software of the equipment. Therefore, PWV measurements obtained by brachial oscillations should not be influenced by body composition changes induced by weight reduction. This was further substantiated by our observation that there was no association between reductions in BMI, body fat % or waist-hip ratio, and improvements in aPWV. Furthermore, results obtained using this method have been shown to correlate with invasive and noninvasive measurements of aPWV39,40 and central blood pressure in normal weight and obese individuals.40,53,54 Thus, Arteriograph-based aPWV measures are likely to provide reliable results independent of body weight. Aortic PWV values are dependent on correct measurements of aortic length. We estimated this as a straight line from the suprapubic notch to the pubic bone as this is considered to provide the most accurate estimate of the aortic pulse wave travel length from the assumed point of reflection around the aortic bifurcation.40 This type of estimation may give a small overestimation of the aortic length as measured by magnetic resonance imaging.55 We avoided variations in aortic length measurement as part of the PWV result by using the initially estimated length also at the follow-up investigation. Although the aortic length may increase over time, such change should be minimal during the study period; moreover, it mainly takes place in the ascending aorta, which is not involved in the calculation of PWV.56 Thus, the alteration of PWV that we have shown can only be explained by an alteration of transit time. The primary outcome in this study was change in aPWV from a baseline value and our results may therefore be influenced by regression towards the mean. However, we believe that this effect is limited because study inclusion and exclusion criteria were not based on aPWV values and all aPWV measures were based on the mean of several individual measurements at each time point. Furthermore, the majority of the subjects improved in PWV over 2 years and the significant difference in aPWV for the entire cohort (P = 0.001) suggest that it is not a random effect. In addition to these considerations, our study included primarily women, which reflects the fact that only a minority of obese subjects undergoing bariatric surgery in Sweden are men. We did not include an obese control group who did not undergo surgery. However, increasing age is a well-known determinant of arterial stiffness, and it is unlikely that spontaneous improvements in PWV would be observed in weight-stable middle-aged individuals over a 2-year observation. Finally, transcriptomic data was only obtained from the subcutaneous WAT depot and we cannot exclude depot-specific differences. In conclusion, our study demonstrates that bariatric surgery improves long-term arterial stiffness. The improvement in aPWV was predicted by subcutaneous adipocyte size and adipose tissue COL4A1 mRNA expression. Although the pathophysiological role of these factors and their value in predicting cardiovascular morbidity/mortality need to be explored in future studies, our results suggest an important link between WAT phenotype and arterial stiffness. SUPPLEMENTARY MATERIAL Supplementary data are available at American Journal of Hypertension online. DISCLOSURE The authors declared no conflict of interest. ACKNOWLEDGMENTS We are grateful for technical assistance provided by Yvonne Widlund, Katarina Hertel, Britt-Marie Leijonhufvud, Elisabeth Dungner, Eva Sjölin, Kerstin Wåhlén, and Gaby Åström. This study was supported by grants from the Swedish Research Council, Novo Nordisk Foundation including the Tripartite Immuno-metabolism Consortium (TrIC) Grant Number NNF15CC0018486 and the MSAM consortium NNF15SA0018346, CIMED (project title: Adipose tissue turnover – mechanisms and clinical impact), Swedish Diabetes Foundation (grant number DIA2016-097), EFSD (project title: A translational approach to identify novel regulators of insulin sensitivity in human adipose tissue), Stockholm County Council (grant numbers 20150517 and K2014-54X-14510-12-5), the Diabetes Research Program at Karolinska Institutet (grant number 2009-1068), Swedish Heart Lung Foundation (grant number 20150423) and The Erling-Persson Family Foundation (grant number 140607). REFERENCES 1. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol  2010; 55: 1318– 1327. Google Scholar CrossRef Search ADS PubMed  2. Mitchell GF, Hwang SJ, Vasan RS, Larson MG, Pencina MJ, Hamburg NM, Vita JA, Levy D, Benjamin EJ. Arterial stiffness and cardiovascular events: the Framingham Heart Study. Circulation  2010; 121: 505– 511. Google Scholar CrossRef Search ADS PubMed  3. Cecelja M, Chowienczyk P. Dissociation of aortic pulse wave velocity with risk factors for cardiovascular disease other than hypertension: a systematic review. Hypertension  2009; 54: 1328– 1336. Google Scholar CrossRef Search ADS PubMed  4. Ashor AW, Lara J, Siervo M, Celis-Morales C, Mathers JC. Effects of exercise modalities on arterial stiffness and wave reflection: a systematic review and meta-analysis of randomized controlled trials. PLoS One  2014; 9: e110034. Google Scholar CrossRef Search ADS PubMed  5. Stefanadis C, Tsiamis E, Vlachopoulos C, Stratos C, Toutouzas K, Pitsavos C, Marakas S, Boudoulas H, Toutouzas P. Unfavorable effect of smoking on the elastic properties of the human aorta. Circulation  1997; 95: 31– 38. Google Scholar CrossRef Search ADS PubMed  6. Safar ME, Czernichow S, Blacher J. Obesity, arterial stiffness, and cardiovascular risk. J Am Soc Nephrol  2006; 17( 4 Suppl 2): S109– S111. Google Scholar CrossRef Search ADS PubMed  7. Webb DR, Khunti K, Silverman R, Gray LJ, Srinivasan B, Lacy PS, Williams B, Davies MJ. Impact of metabolic indices on central artery stiffness: independent association of insulin resistance and glucose with aortic pulse wave velocity. Diabetologia  2010; 53: 1190– 1198. Google Scholar CrossRef Search ADS PubMed  8. Sabbatini AR, Fontana V, Laurent S, Moreno H. An update on the role of adipokines in arterial stiffness and hypertension. J Hypertens  2015; 33: 435– 444. Google Scholar CrossRef Search ADS PubMed  9. Ferreira I, Snijder MB, Twisk JW, van Mechelen W, Kemper HC, Seidell JC, Stehouwer CD. Central fat mass versus peripheral fat and lean mass: opposite (adverse versus favorable) associations with arterial stiffness? The Amsterdam Growth and Health Longitudinal Study. J Clin Endocrinol Metab  2004; 89: 2632– 2639. Google Scholar CrossRef Search ADS PubMed  10. Farb MG, Bigornia S, Mott M, Tanriverdi K, Morin KM, Freedman JE, Joseph L, Hess DT, Apovian CM, Vita JA, Gokce N. Reduced adipose tissue inflammation represents an intermediate cardiometabolic phenotype in obesity. J Am Coll Cardiol  2011; 58: 232– 237. Google Scholar CrossRef Search ADS PubMed  11. Tabara Y, Takahashi Y, Setoh K, Kawaguchi T, Gotoh N, Terao C, Yamada R, Kosugi S, Sekine A, Nakayama T, Matsuda F. Synergistic association of elevated serum free fatty acid and glucose levels with large arterial stiffness in a general population: The Nagahama Study. Metabolism  2016; 65: 66– 72. Google Scholar CrossRef Search ADS PubMed  12. McEniery CM, Wilkinson IB, Johansen NB, Witte DR, Singh-Manoux A, Kivimaki M, Tabak AG, Brunner EJ, Shipley MJ. Nondiabetic glucometabolic status and progression of aortic stiffness: the Whitehall II Study. Diabetes Care  2017; 25: dc16– 1773. 13. Johansen NB, Vistisen D, Brunner EJ, Tabak AG, Shipley MJ, Wilkinson IB, McEniery CM, Roden M, Herder C, Kivimaki M, Witte DR. Determinants of aortic stiffness: 16-year follow-up of the Whitehall II study. PLoS One  2012; 7: 22. 14. Arner P, Bäckdahl J, Hemmingsson P, Stenvinkel P, Eriksson-Hogling D, Näslund E, Thorell A, Andersson DP, Caidahl K, Rydén M. Regional variations in the relationship between arterial stiffness and adipocyte volume or number in obese subjects. Int J Obes (Lond)  2015; 39: 222– 227. Google Scholar CrossRef Search ADS PubMed  15. Lacolley P, Challande P, Osborne-Pellegrin M, Regnault V. Genetics and pathophysiology of arterial stiffness. Cardiovasc Res  2009; 81: 637– 648. Google Scholar CrossRef Search ADS PubMed  16. Petersen KS, Blanch N, Keogh JB, Clifton PM. Effect of weight loss on pulse wave velocity: systematic review and meta-analysis. Arterioscler Thromb Vasc Biol  2015; 35: 243– 252. Google Scholar CrossRef Search ADS PubMed  17. Rider OJ, Tayal U, Francis JM, Ali MK, Robinson MR, Byrne JP, Clarke K, Neubauer S. The effect of obesity and weight loss on aortic pulse wave velocity as assessed by magnetic resonance imaging. Obesity  2010; 18: 2311– 2316. Google Scholar CrossRef Search ADS PubMed  18. Domienik-Karłowicz J, Lisik W, Rymarczyk Z, Dzikowska-Diduch O, Chmura A, Demkow U, Pruszczyk P. The short-term effect of bariatric surgery on non-invasive markers of artery function in patients with metabolic syndrome. Diabetol Metab Syndr  2015; 7: 76. Google Scholar CrossRef Search ADS PubMed  19. Gjevestad E, Hjelmesaeth J, Sandbu R, Nordstrand N. Effects of intensive lifestyle intervention and gastric bypass on aortic stiffness: a 1-year nonrandomized clinical study. Obesity  2015; 23: 37– 45. Google Scholar CrossRef Search ADS PubMed  20. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation  2002; 106: 3143– 3421. PubMed  21. Tomlinson LA. Methods for assessing arterial stiffness: technical considerations. Curr Opin Nephrol Hypertens  2012; 21: 655– 660. Google Scholar CrossRef Search ADS PubMed  22. Hirsch J, Gallian E. Methods for the determination of adipose cell size in man and animals. J Lipid Res  1968; 9: 110– 119. Google Scholar PubMed  23. Arner P, Andersson DP, Thorne A, Wiren M, Hoffstedt J, Naslund E, Thorell A, Ryden M. Variations in the size of the major omentum are primarily determined by fat cell number. J Clin Endocrinol Metab  2013; 98: 2012– 4106. Google Scholar CrossRef Search ADS   24. Rydén M, Hrydziuszko O, Mileti E, Raman A, Bornholdt J, Boyd M, Toft E, Qvist V, Näslund E, Thorell A, Andersson DP, Dahlman I, Gao H, Sandelin A, Daub CO, Arner P. The adipose transcriptional response to insulin is determined by obesity, not insulin sensitivity. Cell Rep  2016; 16: 2317– 2326. Google Scholar CrossRef Search ADS PubMed  25. Cecelja M, Jiang B, Mangino M, Spector TD, Chowienczyk PJ. Association of cross-sectional and longitudinal change in arterial stiffness with gene expression in the Twins UK Cohort. Hypertension  2016; 67: 70– 76. Google Scholar CrossRef Search ADS PubMed  26. Tarasov KV, Sanna S, Scuteri A, Strait JB, Orrù M, Parsa A, Lin PI, Maschio A, Lai S, Piras MG, Masala M, Tanaka T, Post W, O’Connell JR, Schlessinger D, Cao A, Nagaraja R, Mitchell BD, Abecasis GR, Shuldiner AR, Uda M, Lakatta EG, Najjar SS. COL4A1 is associated with arterial stiffness by genome-wide association scan. Circ Cardiovasc Genet  2009; 2: 151– 158. Google Scholar CrossRef Search ADS PubMed  27. Mitchell GF, Verwoert GC, Tarasov KV, Isaacs A, Smith AV, Yasmin, Rietzschel ER, Tanaka T, Liu Y, Parsa A, Najjar SS, O’Shaughnessy KM, Sigurdsson S, De Buyzere ML, Larson MG, Sie MP, Andrews JS, Post WS, Mattace-Raso FU, McEniery CM, Eiriksdottir G, Segers P, Vasan RS, van Rijn MJ, Howard TD, McArdle PF, Dehghan A, Jewell ES, Newhouse SJ, Bekaert S, Hamburg NM, Newman AB, Hofman A, Scuteri A, De Bacquer D, Ikram MA, Psaty BM, Fuchsberger C, Olden M, Wain LV, Elliott P, Smith NL, Felix JF, Erdmann J, Vita JA, Sutton-Tyrrell K, Sijbrands EJ, Sanna S, Launer LJ, De Meyer T, Johnson AD, Schut AF, Herrington DM, Rivadeneira F, Uda M, Wilkinson IB, Aspelund T, Gillebert TC, Van Bortel L, Benjamin EJ, Oostra BA, Ding J, Gibson Q, Uitterlinden AG, Abecasis GR, Cockcroft JR, Gudnason V, De Backer GG, Ferrucci L, Harris TB, Shuldiner AR, van Duijn CM, Levy D, Lakatta EG, Witteman JC. Common genetic variation in the 3’-BCL11B gene desert is associated with carotid-femoral pulse wave velocity and excess cardiovascular disease risk: the AortaGen Consortium. Circ Cardiovasc Genet  2012; 5: 81– 90. Google Scholar CrossRef Search ADS PubMed  28. Reference Values for Arterial Stiffness’ Collaboration, Mattace-Raso FHA, Verwoert GC, Wittemana JC, Wilkinson I, Cockcroft J, McEniery C, Yasmin, Laurent S, Boutouyrie P, Bozec E, Hansen TW, Torp-Pedersen C, Ibsen H, Jeppesen J, Vermeersch SJ, Rietzschel E, De Buyzere M, Gillebert TC, Van Bortel L, Segers P, Vlachopoulos C, Aznaouridis C, Stefanadis C, Benetos A, Labat C, Lacolley P, Stehouwer C, Nijpels G, Dekker JM, Stehouwer C, Ferreira I, Twisk JW, Czernichow S, Galan P, Hercberg S, Pannier B, Guérin A, London G, Cruickshank JK, Anderson SG, Paini A, Agabiti Rosei E, Muiesan ML, Salvetti M, Filipovsky J, Seidlerova J, Dolejsova M. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J  2010; 31: 2338– 2350. Google Scholar CrossRef Search ADS PubMed  29. Guerin AP, Blacher J, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness attenuation on survival of patients in end-stage renal failure. Circulation  2001; 103: 987– 992. Google Scholar CrossRef Search ADS PubMed  30. Arner E, Ryden M, Arner P. Tumor necrosis factor alpha and regulation of adipose tissue. N Engl J Med  2010; 362: 1151– 1153. Google Scholar CrossRef Search ADS PubMed  31. Virtue S, Vidal-Puig A. Adipose tissue expandability, lipotoxicity and the Metabolic Syndrome–an allostatic perspective. Biochim Biophys Acta  2010; 1801: 338– 349. Google Scholar CrossRef Search ADS PubMed  32. Hoffstedt J, Arner E, Wahrenberg H, Andersson DP, Qvisth V, Löfgren P, Rydén M, Thörne A, Wirén M, Palmér M, Thorell A, Toft E, Arner P. Regional impact of adipose tissue morphology on the metabolic profile in morbid obesity. Diabetologia  2010; 53: 2496– 2503. Google Scholar CrossRef Search ADS PubMed  33. Timpl R. Structure and biological activity of basement membrane proteins. Eur J Biochem  1989; 180: 487– 502. Google Scholar CrossRef Search ADS PubMed  34. Wang MLE. Central Arterial Aging: Humans to Molecules Hypertension in the Elderly , Vol. 23. Amsterdam, The Netherlands: Elsevier, 2006. 35. Tsilibary EC, Charonis AS, Reger LA, Wohlhueter RM, Furcht LT. The effect of nonenzymatic glucosylation on the binding of the main noncollagenous NC1 domain to type IV collagen. J Biol Chem  1988; 263: 4302– 4308. Google Scholar PubMed  36. Reggio S, Rouault C, Poitou C, Bichet JC, Prifti E, Bouillot JL, Rizkalla S, Lacasa D, Tordjman J, Clément K. Increased basement membrane components in adipose tissue during obesity: links with TGFβ and metabolic phenotypes. J Clin Endocrinol Metab  2016; 101: 2578– 2587. Google Scholar CrossRef Search ADS PubMed  37. Lakatta EG. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: Part III: cellular and molecular clues to heart and arterial aging. Circulation  2003; 107: 490– 497. Google Scholar CrossRef Search ADS PubMed  38. Sakurai M, Yamakado T, Kurachi H, Kato T, Kuroda K, Ishisu R, Okamoto S, Isaka N, Nakano T, Ito M. The relationship between aortic augmentation index and pulse wave velocity: an invasive study. J Hypertens  2007; 25: 391– 397. Google Scholar CrossRef Search ADS PubMed  39. Baulmann J, Schillings U, Rickert S, Uen S, Düsing R, Illyes M, Cziraki A, Nickering G, Mengden T. A new oscillometric method for assessment of arterial stiffness: comparison with tonometric and piezo-electronic methods. J Hypertens  2008; 26: 523– 528. Google Scholar CrossRef Search ADS PubMed  40. Horváth IG, Németh A, Lenkey Z, Alessandri N, Tufano F, Kis P, Gaszner B, Cziráki A. Invasive validation of a new oscillometric device (Arteriograph) for measuring augmentation index, central blood pressure and aortic pulse wave velocity. J Hypertens  2010; 28: 2068– 2075. Google Scholar CrossRef Search ADS PubMed  41. Ikonomidis I, Ntai K, Kadoglou NP, Papadakis I, Kornelakis M, Tritakis V, Varoudi M, Papadima T, Triantafyllidi H, Parissis J, Paraskevaidis I, Lekakis J. The evaluation of pulse wave velocity using Arteriograph and Complior apparatus across multiple cohorts of cardiovascular-related diseases. Int J Cardiol  2013; 168: 4890– 4892. Google Scholar CrossRef Search ADS PubMed  42. Rajzer MW, Wojciechowska W, Klocek M, Palka I, Brzozowska-Kiszka M, Kawecka-Jaszcz K. Comparison of aortic pulse wave velocity measured by three techniques: Complior, SphygmoCor and Arteriograph. J Hypertens  2008; 26: 2001– 2007. Google Scholar CrossRef Search ADS PubMed  43. Jatoi NA, Mahmud A, Bennett K, Feely J. Assessment of arterial stiffness in hypertension: comparison of oscillometric (Arteriograph), piezoelectronic (Complior) and tonometric (SphygmoCor) techniques. J Hypertens  2009; 27: 2186– 2191. Google Scholar CrossRef Search ADS PubMed  44. Ring M, Eriksson MJ, Zierath JR, Caidahl K. Arterial stiffness estimation in healthy subjects: a validation of oscillometric (Arteriograph) and tonometric (SphygmoCor) techniques. Hypertens Res  2014; 37: 999– 1007. Google Scholar CrossRef Search ADS PubMed  45. Cruickshank JK, Silva MJ, Molaodi OR, Enayat ZE, Cassidy A, Karamanos A, Read UM, Faconti L, Dall P, Stansfield B, Harding S. Ethnic differences in and childhood influences on early adult pulse wave velocity: the determinants of adolescent, now Young Adult, Social Wellbeing, and Health Longitudinal Study. Hypertension  2016; 67: 1133– 1141. Google Scholar CrossRef Search ADS PubMed  46. Rezai MR, Wallace AM, Sattar N, Finn JD, Wu FC, Cruickshank JK. Ethnic differences in aortic pulse wave velocity occur in the descending aorta and may be related to vitamin D. Hypertension  2011; 58: 247– 253. Google Scholar CrossRef Search ADS PubMed  47. Fodor L, Premuzic V, Ivkovic V, Perkov D, Laganovic M, Vrkic TZ, Dika Z, Zivko M, Jelakovic B. Arterial stiffness in atherosclerotic renovascular hypertension. J Hypertens  2014; 32: 2238– 2245. Google Scholar CrossRef Search ADS PubMed  48. Tarnoki AD, Tarnoki DL, Stazi MA, Medda E, Cotichini R, Nisticò L, Fagnani C, Lucatelli P, Boatta E, Zini C, Fanelli F, Baracchini C, Meneghetti G, Osztovits J, Jermendy G, Préda I, Kiss RG, Metneki J, Horvath T, Karlinger K, Racz A, Lannert A, Molnar AA, Littvay L, Garami Z, Berczi V, Schillaci G. Heritability of central blood pressure and arterial stiffness: a twin study. J Hypertens  2012; 30: 1564– 1571. Google Scholar CrossRef Search ADS PubMed  49. Tarnoki A, Tarnoki DL, Godor E, Littvay L, Horvath T, Jermendy A, Merkely B, Jermendy G, Maurovich-Horvat P. 2d.05: relationship of coronary atherosclerosis with arterial stiffness and central systolic blood pressure. J Hypertens  2015; 33: 74. 50. Faconti L, Silva MJ, Molaodi OR, Enayat ZE, Cassidy A, Karamanos A, Nanino E, Read UM, Dall P, Stansfield B, Harding S, Cruickshank KJ. Can arterial wave augmentation in young adults help account for variability of cardiovascular risk in different British ethnic groups? J Hypertens  2016; 34: 2220– 2226. Google Scholar CrossRef Search ADS PubMed  51. Osman MW, Leone F, Nath M, Khalil A, Webb DR, Robinson TG, Mousa HA. Diurnal variation and repeatability of arterial stiffness and cardiac output measurements in the third trimester of uncomplicated pregnancy. J Hypertens  2017; 35: 2436– 2442. Google Scholar CrossRef Search ADS PubMed  52. Tarnoki AD, Tarnoki DL, Bogl LH, Medda E, Fagnani C, Nisticò L, Stazi MA, Brescianini S, Lucatelli P, Boatta E, Zini C, Fanelli F, Baracchini C, Meneghetti G, Osztovits J, Jermendy G, Kiss RG, Preda I, Karlinger K, Lannert A, Molnar AA, Littvay L, Garami Z, Berczi V, Pucci G, Baffy G, Schillaci G, Pietiläinen KH. Association of body mass index with arterial stiffness and blood pressure components: a twin study. Atherosclerosis  2013; 229: 388– 395. Google Scholar CrossRef Search ADS PubMed  53. Rossen NB, Laugesen E, Peters CD, Ebbehøj E, Knudsen ST, Poulsen PL, Bøtker HE, Hansen KW. Invasive validation of arteriograph estimates of central blood pressure in patients with type 2 diabetes. Am J Hypertens  2014; 27: 674– 679. Google Scholar CrossRef Search ADS PubMed  54. Echeverri D, Pizano A, Cabrales J, Moreno K. Validation of central and peripheral non-invasive hemodynamic variables using an oscillometric method. High Blood Press Cardiovasc Prev  2017; 28: 017– 0238. 55. Rezai MR, Cowan BR, Sherratt N, Finn JD, Wu FC, Cruickshank JK. A magnetic resonance perspective of the pulse wave transit time by the Arteriograph device and potential for improving aortic length estimation for central pulse wave velocity. Blood Press Monit  2013; 18: 111– 118. Google Scholar CrossRef Search ADS PubMed  56. Sugawara J, Hayashi K, Yokoi T, Tanaka H. Age-associated elongation of the ascending aorta in adults. JACC Cardiovasc Imaging  2008; 1: 739– 748. Google Scholar CrossRef Search ADS PubMed  © American Journal of Hypertension, Ltd 2017. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

American Journal of HypertensionOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from Google Scholar, PubMed
Create lists to organize your research
Export lists, citations
Access to DeepDyve database
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off