Skip to main content
NCBI home page
Chest logoLink to Chest
. 2020 Jul 24;159(1):302–310. doi: 10.1016/j.chest.2020.07.038

BMI Is Causally Associated With Pulmonary Artery Pressure But Not Hemodynamic Evidence of Pulmonary Vascular Remodeling

Timothy E Thayer a, Rebecca T Levinson a, Shi Huang b, Tufik Assad c, Eric Farber-Eger d, Quinn S Wells a, Jonathan D Mosley e, Evan L Brittain a,∗
PMCID: PMC8008481  PMID: 32712226

Abstract

Background

There is an unclear relationship of obesity to the pathogenesis and severity of pulmonary arterial hypertension (PAH) and pulmonary venous hypertension (PVH).

Research Question

Is BMI casually associated with pulmonary artery pressure (PAP) and/or markers of pulmonary vascular remodeling?

Study Design and Methods

The study design was a two-sample inverse-variance weighted Mendelian randomization. We constructed two BMI genetic risk scores from genome-wide association study summary data and deployed them in nonoverlapping cohorts of subjects referred for right heart catheterization (RHC) or echocardiography. A BMI highly polygenic risk score (hpGRS) optimally powered to detect shared genetic architecture of obesity with other traits was tested for association with RHC parameters including markers of pulmonary vascular remodeling. A BMI strict genetic risk score (sGRS) composed of high-confidence genetic variants was used for Mendelian randomization analyses to assess if higher BMI causes higher PAP.

Results

Among all subjects, both directly measured BMI and hpGRS were positively associated with pulmonary arterial pressures but not markers of pulmonary vascular remodeling. Categorical analyses revealed BMI and hpGRS were associated with PVH but not PAH. Mendelian randomization of the sGRS supported that higher BMI is causal of higher systolic pulmonary artery pressure (sPAP). Sensitivity analyses showed sPAP-BMI sGRS relationship was preserved when either individuals with PAH or PVH were excluded. In the echocardiographic cohort, BMI and hpGRS were positively associated with estimated PAP and markers of left heart remodeling.

Interpretation

BMI is a modifier of pulmonary hypertension severity in both PAH and PVH but is only involved in the pathogenesis of PVH.

Key Words: echocardiogram, mendelian randomization, obesity, pulmonary hypertension, right heart catheterization

Abbreviations: GRS, genetic risk score; GWAS, genome-wide association study; hpGRS, highly polygenic risk score; MEGA, Multi-Ethnic Global Array; PAH, pulmonary arterial hypertension; PH, pulmonary hypertension; PVH, pulmonary venous hypertension; RHC, right heart catheterization; sGRS, strict genetic risk score; sPAP, systolic pulmonary artery pressure

Take-home Point.

It was unknown if BMI is causally associated with pulmonary artery pressure and/or pulmonary vascular remodeling. Mendelian randomization analyses support that higher BMI causes elevated pulmonary artery pressure but not pulmonary vascular remodeling.

Our group and others have described the overlap of pulmonary hypertension (PH) and metabolic dysfunction, but a biological link between obesity and PH has not been established in humans.1,2 Results of previous studies assessing the relationship of obesity and pulmonary arterial hypertension (PAH) have been mixed, with some suggesting that obesity is a risk factor for subtypes of PAH.3, 4, 5 Observational studies of pulmonary venous hypertension (PVH), by contrast, revealed obesity as a risk factor for heart failure and, by extension, PVH.5, 6, 7 It is unknown whether the obesity-PVH association is mediated by the effects of increased body mass or by biological mechanisms that independently modulate each measure arising from a shared genetic architecture of the two diseases.8

Based on published observational studies, we hypothesized that both BMI and genetic risk for obesity would be associated with PVH but not PAH. We built two BMI genetic risk scores: one highly polygenic risk score (hpGRS) with maximal power to detect shared genetic architecture of BMI and PH and a second for Mendelian randomization analyses to assess if obesity causes higher pulmonary arterial pressure. We used two large, nonoverlapping genotyped electronic health record cohorts of subjects referred for right heart catheterization (RHC) or echocardiography to test for the association of directly measured BMI and BMI genetic risk scores with invasive hemodynamics and echocardiographic parameters.

Methods

Study Subjects and Clinical Data

All analyses were carried out on data obtained from Vanderbilt University Medical Center’s deidentified electronic health record. Available data were obtained as a part of routine clinical practice, and the study was approved by Vanderbilt’s institutional review board. Methods for cohort creation have been published previously.9, 10, 11 The population referred for RHC consisted of individuals with existing genotyping on the Illumina Multi-Ethnic Global Array (MEGA) platform. The echocardiogram cohort consisted of individuals referred for echocardiogram as part of routine clinical practice who also had MEGA genotyping available. Any individuals in the RHC cohort were excluded from the echocardiographic cohort. If multiple echocardiograms were available, data from the first echocardiogram were used. Individuals in each cohort were required to be of European ancestry by principal components and related individuals (pi-hat > 0.05) were excluded using PLINK 1.9.12

Genetic Risk Scores for BMI and Height

Genetic risk scores have emerged as powerful tools with multiple applications. A genetic risk score is built from existing genome-wide association study (GWAS) summary statistics, which results in a single numerical predictor that distills the genetic risk of an individual toward a given trait based on common genetic variants. When horizontal pleiotropy is not present, genetic risk scores can be used in Mendelian randomization experiments.13 Mendelian randomization takes advantage of the meiotic process to randomize delivery genetic variants to individuals at conception as a sort of randomized controlled trial. When genetic variants are known to carry risk of a given trait/exposure, Mendelian randomization experiments avoid the problem of reverse causality and effectively determine direction of effect. There has been a trend toward producing hpGRSs including many thousands of genetic variants.14,15 hpGRSs maximize predictive power for a given trait but are uncommonly used for Mendelian randomization studies because the pleiotropic effects of low confidence genetic variants undermine the assumptions of most Mendelian randomization approaches. Still, hpGRSs can be used to test whether there is shared genetics between the predicted trait and another.16, 17, 18, 19

Two distinct BMI genetic risk scores were used for analyses: a strict genetic risk score (sGRS) for Mendelian randomization analyses and an hpGRS that was better powered to identify associations. We used publicly available summary statistics from a published BMI GWAS of approximately 700,000 individuals from the Genetic Investigation of Anthropometric Traits consortium as the basis for creating the genetic risk scores.20 In each cohort, genetic risk scores were centered so that a score of 0 represented the mean genetic risk score (GRS) and each SD from the mean was valued at 1 (or −1). Further details of the production of the genetic risk scores are available in e-Appendix 1.

As a negative control, hpGRSs for height were derived from a GWAS on height from the Genetic Investigation of Anthropometric Traits consortium in the same process previously mentioned, which is further described in e-Appendix 1.20

Association Studies of BMI Genetic Risk Score With Hemodynamic and Structural Variables Assessed by RHC or Echocardiography

Demographics, hemodynamics, and echocardiographic data were extracted from clinical reports available in Vanderbilt’s deidentified electronic health record as previously described.9,21 PAH and PVH were defined by the European Society of Cardiology/European Respiratory Society 2015 guidelines.22

Statistical Analyses

Summary statistics, such as median (interquartile range) and percent (No.), were presented for demographics, RHC parameters, and echocardiographic variables. For assessment of differences in summary statistics among groups by BMI genetic risk score quartile, all continuous variable groups were compared using the Kruskal-Wallis rank sum test and categorical variables were compared using the χ2 test.

BMI and BMI hpGRS were tested for association by regression models with clinically collected RHC and echocardiographic values, adjusting for potential confounders, including age, sex, and principal components 1 through 3. Principal components account for population stratification because of continental ancestry and were created using PLINK1.9.12,23 The genotyping batch was also adjusted for in the models with echocardiographic values as the dependent variables to account for potential population differences between batches.

Most outcome variables had a highly skewed distribution which violates the normality assumption of linear regression modeling; therefore, we modeled continuous outcomes using the ordinal regression model. Ordinal regression makes no assumption about the distribution of outcomes. We used logistic distribution function (ie, proportional odds model, logit link function) for this study, and the ordinal cumulative probability models are stated as follows: Prob [Y ≥ y | X].

The possible nonlinear relationship between BMI and BMI genetic risk score and outcomes was examined by using a restricted cubic spline with three knots. If the nonlinear terms of BMI and BMI genetic risk scores were not statistically significant, we then removed the nonlinear terms from the model to simplify interpretations. The BMI and BMI hpGRS interquartile range ORs with their 0.95 confidence limits were reported to represent the magnitude of their independent effects on outcomes. To give clinical context for an effect size, we estimated the adjusted mean outcome levels for a patient with the 25th and 75th percentile of BMI and BMI hpGRS. Based on the cumulative probability ordinal model, the predicted mean of Y|X was estimated by computing the following:∑i=1nyiProb[Y=yi|X].24

For binary variables (PAH and PVH), logistic regression was used to adjust for age, sex, and principal components 1 through 3. When testing the association of PAH with BMI genetic risk score (or BMI), subjects with PVH were excluded and vice versa. Nonlinear terms were considered for the relationship between BMI and BMI genetic risk score with PAH/PVH using a restricted cubic spline with three knots.

For Mendelian randomization, our primary analysis was a two-sample inverse-variance weighted Mendelian randomization using both local individual-level data and publicly available summary-level data. Inverse-weighted variance random effect modeling was performed using the Mendelian Randomization R package as previously described.13,25,26 The combined effect of the genetic variants in the BMI sGRS were used as an instrument variable to test if BMI causally affects the outcomes of interest: systolic pulmonary artery pressure (sPAP) in the RHC cohort, and left ventricular posterior wall thickness and left atrial diameter in the echocardiographic cohort. In follow-up analyses, individuals with either PVH of PAH were excluded from analysis to test if either phenotype was driving the association of BMI with systolic pulmonary arterial pressure. Further details of Mendelian randomization and sensitivity analyses are available in e-Appendix 1.

This study was approved by the Vanderbilt University Medical Center institutional review board (No. 140544).

Results

We identified 1,043 subjects with existing MEGA genotyping data who were referred for RHC, unrelated, and of European ancestry (52% men; age, 60 ± 14 years). In this population, 12% had PAH, 40% had PVH, 6% had PH because of lung disease, and 42% did not have PH. Demographics, comorbidities, and laboratory values of the RHC cohort stratified by BMI genetic risk score quartile are displayed in Table 1.

Table 1.

Demographics, Comorbidities, and Laboratory Values of the Right Heart Catheterization Cohort Stratified by BMI Genetic Risk Score Quartile

Variable Quartile 1 (n = 261) Quartile 2 (n = 261) Quartile 3 (n = 261) Quartile 4 (n = 260) P Value
Male 123 (47.1) 148 (56.7) 138 (52.9) 136 (52.3) .182
Age, y 63.33 (52.55-71.62) 60.97 (52.10-68.90) 61.48 (53.34-69.21) 60.41 (50.56-70.49) .125
Weight, kg 77.62 (65.32-89.47) 84.76 (73.48-99.34) 85.55 (72.95-100.20) 88.45 (74.54-102.25) < .001
BMI, kg/m2 26.34 (23.12-30.33) 28.79 (25.15-33.61) 29.00 (25.29-33.48) 29.44 (26.41-34.82) < .001
Atrial fibrillation 88 (33.7) 90 (34.5) 85 (32.6) 85 (32.7) .962
COPD 29 (11.1) 42 (16.1) 33 (12.6) 41 (15.8) .278
Coronary artery disease 194 (74.3) 195 (74.7) 200 (76.6) 188 (72.3) .731
Diabetes mellitus, type 2 96 (36.8) 100 (38.3) 114 (43.7) 115 (44.2) .206
Heart failure 131 (50.2) 147 (56.3) 135 (51.7) 149 (57.3) .287
OSA 22 (8.4) 32 (12.3) 36 (13.8) 32 (12.3) .263
Systemic hypertension 216 (82.8) 227 (87.0) 214 (82.0) 212 (81.5) .322
Brain natriuretic peptide 372.00 (153.00-973.00) 317.00 (102.00-707.00) 333.00 (122.00-643.00) 258.00 (107.00-788.00) .61
Creatinine 1.00 (0.81-1.30) 1.00 (0.84-1.27) 1.03 (0.83-1.30) 1.02 (0.84-1.30) .10
HDL 43.00 (32.00-54.00) 37.00 (31.00-51.00) 40.00 (32.00-49.00) 39.00 (30.50-47.00) .104
Hemoglobin 13.00 (11.57-14.50) 13.20 (11.60-14.50) 12.90 (11.50-14.40) 12.85 (11.50-14.20) .40
HgbA1c 5.80 (5.40-6.40) 5.90 (5.40-6.70) 5.90 (5.50-6.60) 6.00 (5.50-6.90) .003
LDL 82.50 (60.75-115.25) 87.00 (62.00-116.00) 88.00 (66.00-118.00) 91.00 (62.00-114.00) .776
Sodium 138.00 (136.00-141.00) 138.00 (136.00-140.00) 139.00 (137.00-140.00) 138.00 (136.00-141.00) .892
Triglycerides 103.00 (75.00-146.00) 121.00 (87.25-176.75) 121.50 (85.00-182.00) 121.00 (86.00-177.00) .027
Open in a new tab

Values are median (25th percentile-75th percentile), No. (%), or as otherwise indicated. All continuous variable groups were compared using the Kruskal-Wallis rank sum test. Categorical variables were compared using χ2 testing.

HDL = high density lipoprotein; HgbA1c = glycosylated hemoglobin.

Association of BMI With RHC Parameters

BMI obtained at the time of RHC was positively associated mean pulmonary artery pressure and mean pulmonary capillary wedge pressure (Fig 1A). The figure displays predicted differences in hemodynamics between the 75th vs 25th percentile BMI (33 and 25 kg/m2, respectively, in the cohort). The known association of BMI and systemic BP was recapitulated in the cohort. BMI was also associated with higher systolic and diastolic pulmonary arterial pressures, pulmonary wedge pressure, and right atrial pressure. BMI was negatively associated with pulmonary vascular resistance but not other markers of pulmonary vascular remodeling: transpulmonary gradient and diastolic pressure gradient. Full tabular results are shown in e-Table 1. To interrogate if the relationship of BMI to sPAP was similar in PAH compared with the rest of the cohort, a spline graph modeling sPAP-BMI stratified by PAH status was created (e-Fig 1). Based on linear regression modeling, there was no difference in the slopes of the relationship of sPAP to BMI in PAH vs non-PAH (P = .78).

Figure 1.

Figure 1

Open in a new tab

A-B, BMI and BMI genetic risk score were positively associated with systemic and pulmonary pressures but not markers of pulmonary vascular remodeling. A, Displayed values are ORs and 95% CIs comparing individuals at the 75th percentile of the BMI (BMI of 33 kg/m2 in this cohort) with those at the 25th percentile (BMI of 25 kg/m2). An OR > 1 represents that individuals at the 75th percentile were predicted to have a higher measurement than individuals at the 25th percentile. B, Displayed values are ORs and 95% CIs comparing individuals at the 75th percentile of the BMI genetic risk score with those at the 25th percentile. All values in blue have a P value < .05 by ordinal regression adjusted for age at time of right heart catherization, sex, and race by principal components 1 through 3. DPG = diastolic pressure gradient; MAP = mean arterial pressure; mPAP = mean pulmonary artery pressure; mPWP = mean pulmonary wedge pressure; PVR = pulmonary vascular resistance; RAP = right atrial pressure; SBP = systolic BP; sPAP = systolic pulmonary artery pressure; TPG = transpulmonary gradient.

Association of BMI hpGRS With RHC Parameters

The BMI hpGRS with the strongest association with BMI in the cohort contained 31,901 independent genetic variants with P ≤ .20 in the published BMI GWAS. Scatterplot of BMI-BMI hpGRS is available in e-Figure 2 (R = 0.25; P < .0001). Multiple BMI GRSs were validated against BMI and were associated with pulmonary artery and pulmonary capillary wedge pressures (e-Fig 3).

The BMI hpGRS was positively associated with multiple hemodynamic parameters including pulmonary artery pressures and pulmonary wedge pressure (Fig 1B). There was no association between the BMI hpGRS with pulmonary vascular resistance or other hemodynamic metrics of pulmonary vascular remodeling; therefore, those variables were not included in the Mendelian randomization analyses. Full tabular results are shown in e-Table 2. A height GRS was used as a control and was not associated with hemodynamic parameters (e-Fig 4, e-Table 3).

Association of BMI hpGRS With PH Subgroups

Next, BMI and the BMI hpGRS were tested for association with PAH and PVH. The BMI hpGRS was used over the BMI sGRS to maximize the power to detect shared genetic architecture of BMI with each PH etiology. Both BMI and BMI genetic risk score were associated with PVH but not PAH (Fig 2). The nonlinear BMI and BMI hpGRS terms were not significantly associated with PAH or PVH (P > .30 for all).

Figure 2.

Figure 2

Open in a new tab

Both BMI and BMI GRS were associated with PVH but not PAH. Displayed values are ORs and 95% CIs comparing individuals at the 75th percentile of either BMI or BMI GRS with those at the 25th percentile. BMI at the 25th percentile was 25 kg/m2 and 75th percentile was 33 kg/m2. Displayed P values were calculated by logistic regression adjusted for age at time of right heart catherization, sex, and principal components 1 through 3. BMI GRS = highly polygenic BMI genetic risk score; PAH = pulmonary arterial hypertension; PVH = pulmonary venous hypertension

Mendelian Randomization Analysis of BMI sGRS and Pulmonary Arterial Pressure

The BMI sGRS, composed of 64 genetic variants, was associated with measured BMI in the RHC cohort, although, as expected, with lower predictive power than the BMI hpGRS (R = 0.11; P < .001). The inverse-variance weighted model supported that higher BMI caused higher sPAP with an estimate of 1.1 and SE of 0.4 (P = .004) (Fig 3A). To test if either PVH or PAH were driving the relationship of BMI to pulmonary pressure, each group was selectively removed and the Mendelian randomization analyses were repeated. When either individuals with PVH (Fig 3B) or PAH (Fig 3C) were excluded, the relationship of BMI to pulmonary artery pressure was preserved, suggesting that BMI influences pulmonary arterial pressure through a non-disease-specific mechanism.

Figure 3.

Figure 3

Open in a new tab

A-C, Elevated BMI caused increased sPAP as assessed from inverse-variance weighted mendelian randomization analysis in the entire right heart catheterization (RHC) cohort (A), RHC cohort with subjects with PVH removed (B), or RHC cohort with subjects with PAH removed (C). Each point represents a single nucleotide polymorphism (SNP) from the strict BMI genetic risk score. Each point is plotted at the beta value for the association of the SNP with BMI (x axis) and sPAP (y axis). Beta values and SDs for sPAP were determined for each SNP by ordinal regression modeling adjusted for age, sex, and race by principal components. Because ordinal, not linear, regression was used to model the effect of each genetic variant on sPAP, the estimates do not reflect the relationship of BMI to sPAP in true units but do support causality. Error lines represent SDs. P values and regression lines were derived from inverse-variance weighted modeling summarizing the association of BMI and sPAP. See Figure 1 and 2 legends for expansion of other abbreviations.

Sensitivity analyses carried out with alternative Mendelian randomization approaches produced similar estimates for the causal effect of BMI on sPAP (results available in e-Appendix 1).

Association of BMI and BMI Genetic Risk Score With Cardiac Structure and Function

We used a previously genotyped echocardiogram cohort for the following two purposes: (1) to test for replication of the association of the BMI hpGRS with pulmonary pressure measurements in a nonoverlapping cohort, and (2) to assess the relationship of BMI genetics with left heart structure and function. We identified 5,038 subjects referred for echocardiogram with existing genotyping data (49% men; age, 64 ± 13 years; 36.7% obese; 29.7% with PH by sPAP ≥ 40 mm Hg). As in the invasive hemodynamic cohort, BMI was associated with higher sPAP and markers of left ventricular remodeling including left atrial diameter and left ventricular posterior wall thickness (e-Fig 5A). Full tabular results available in e-Table 4. Even when correcting for left atrial diameter, BMI was positively associated with sPAP (P < .0001).

The BMI hpGRS in the echocardiographic cohort used 31,770 available genetic variants (99.6% of genetic variants used in the invasive hemodynamics cohort). The BMI hpGRS was similarly associated with BMI in this cohort as in the catheterization cohort (R = 0.21; P < .0001). The BMI hpGRS was associated with estimated right ventricular systolic pressure and markers of left-sided remodeling including larger left atrial diameter and increased left ventricular posterior wall thickness (e-Fig 5B). The BMI hpGRS was not associated with left ventricular ejection fraction. Full tabular results are available in e-Table 5.

Next, we used the genetic variants from the BMI sGRS to perform a Mendelian randomization experiment assessing the potential causal relationship of higher BMI with left heart remodeling. Inverse-variance weighted modeling revealed that increased BMI caused left heart remodeling as evidenced by higher left ventricle posterior wall diameter and larger left atrial diameter (e-Fig 6). Full results are available in e-Appendix 1.

Discussion

The reported intersection of metabolic disease and PH prompted us to investigate the relationships of BMI and BMI genetic architecture with pulmonary hemodynamics. Both BMI and the BMI hpGRS were associated with higher pulmonary arterial pressures but not markers of pulmonary vascular remodeling. Consistent with these hemodynamic associations, BMI and the BMI hpGRS were associated with PVH, but not PAH. Utilization of a BMI sGRS in Mendelian randomization analyses supported that higher BMI caused higher pulmonary arterial pressure. Repeat Mendelian randomization excluding individuals with PVH demonstrated that BMI likely influences pulmonary artery pressure through mechanisms distinct from left heart disease. The association of the BMI hpGRS with echocardiographic estimates of pulmonary artery pressure was replicated in a separate cohort. These findings are important because (1) they robustly demonstrate the lack of a clinical or genetic association between obesity and pulmonary vascular disease in humans; and (2) they support that higher BMI causes increased pulmonary arterial pressure, which may have treatment implications for both PAH and PVH.

Obesity has been linked to metabolic dysfunction across a broad spectrum of diseases, but metabolic dysfunction in PAH may not be driven by obesity. Despite the strong association of obesity and insulin resistance in the general population, Pugh et al27 found that insulin resistance in patients with PAH was not correlated with BMI. Therefore, the increased insulin resistance found in the PAH population (which is associated with worse survival) may not be influenced by obesity.28 Metabolic derangement in PAH may be driven predominantly by right ventricular failure leading to altered energy substrate utilization and systemic inflammation.29, 30, 31 Our findings support the notion that targeting weight loss in PAH is unlikely to reverse pulmonary vascular remodeling, but may result in decreased pulmonary artery pressure.

The obesity paradox has also been described in both PAH and heart failure in which obese individuals have a lower mortality than leaner individuals.32, 33, 34 Obese individuals tend to be younger than nonobese individuals in these cohorts which highlights that the obesity paradox may originate from an earlier disease onset of PH in obese individuals. Right ventricular dysfunction is the predominant driver of symptoms and mortality in PAH, and right ventricle dysfunction in PAH occurs because of prolonged exposure to elevated pulmonary artery pressure.35,36 Therefore, it is possible that the elevated pulmonary pressure caused by higher BMI unmasks symptoms of right ventricular dysfunction in obese subjects, which leads to earlier diagnosis of PH.

Previous studies have determined, by Mendelian randomization, that higher BMI is associated with systemic hypertension and heart failure, but no studies have tested the association of BMI genetics with PH.37, 38, 39 The hpGRS analyses confirmed that there is no shared genetic architecture between BMI and pulmonary vascular remodeling. The findings of our Mendelian randomization experiment support that higher BMI causes higher systolic pulmonary arterial pressure. In parallel with our results, Frank et al5 recently demonstrated that BMI was associated with pulmonary pressure in both PAH and PVH in a clinical cohort. Our analyses replicate this finding and, most importantly, provide evidence for causality. Our results also fit with a meta-analysis which concluded that weight loss led to decreased pulmonary artery pressure.40 Given our Mendelian randomization results were recapitulated even when individuals with PVH were excluded, it is likely that BMI influences pulmonary artery pressure through both left heart remodeling and nonleft heart remodeling mechanisms.

Our Mendelian randomization analyses in the echocardiographic cohort revealed that BMI had a positive causal relationship with left ventricular remodeling which supports a likely mechanistic link between BMI and risk of PVH. These findings are consistent with prior studies that identified that obesity is causally associated with left ventricular mass and heart failure with preserved ejection fraction.41,42 Our echocardiographic cohort analyses add to the evidence that obesity is involved in the pathogenesis of PVH.

Our study has important limitations. The generalizablity of our findings to populations of non-European descent is limited because the cohorts used were all of European descent. The predictive power of genetic risk scores are optimized by using GWAS and target population of the same ancesteral background.43 Moveover, the invasive hemodynamic cohort was enriched in individuals with heart failure and PH because it consisted of patients who were referred for a clinically indicated RHC. As such, our findings may not be broadly generalizable to the community; however, our findings are highly relevant to patients seeking care with cardiopulmonary symptoms. Also, BMI, hemodynamic data, and echocardiographic data were obtained as part of routine clinical care and not according to a standandized protocol or interpretation by a central core laboratory. An extension of the nonstandardized data acquisition limitation is that accurate measurement of pulmonary wedge pressure is difficult in obese individuals because of extensive respiratory variation.44 It has previously been demonstrated that pulmonary artery pulse pressure is usually overestimated in obesity by approximately 1 mm Hg.45 Given the difference in mean systolic artery pressure between the first and fourth BMI quartile in the cohort was > 5 mm Hg, measurement error could only account for a small amount of the signal in this cohort.

Interpretation

Our findings suggest that higher BMI is not associated with pulmonary vascular remodeling but is causally associated with higher pulmonary artery pressure.

Acknowledgments

Author contributions: T. E. T. takes responsibility for the content of the manuscript, including the data and analysis. T. E. T., R. T. L., S. H., T. A., E. F.-E., Q. S. W., J. D. M., and E. L. B. were involved in the experimental design, execution, and analyses. T. E. T. and E. L. B. prepared the manuscript, and all authors gave critical review of the manuscript.

Financial/nonfinancial disclosures: None declared.

Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.

Additional information: The e-Appendix, e-Figures, and e-Tables can be found in the Supplemental Materials section of the online article.

Footnotes

FUNDING/SUPPORT: This work was supported by the National Institutes of Health (NIH) [Grants HL146588-01, HL146588-01S1, HL125212-01]. The samples and/or dataset(s) used for the analyses described were obtained from Vanderbilt University Medical Center’s BioVU which is supported by numerous sources: institutional funding, private agencies, and federal grants. These include the NIH funded Shared Instrumentation Grant S10OD017985 and S10RR025141; and CTSA grants UL1TR002243, UL1TR000445, and UL1RR024975. Genomic data are also supported by investigator-led projects that include U01HG004798, R01NS032830, RC2GM092618, P50GM115305, U01HG006378, U19HL065962, R01HD074711; and additional funding sources listed at https://victr.vumc.org/biovu-funding/.

Supplementary Data

e-Online Data
mmc1.pdf (701.5KB, pdf)

References

  • 1.Ayinapudi K., Singh T., Motwani A., Le Jemtel T.H., Oparil S. Obesity and pulmonary hypertension. Curr hypertens Rep. 2018;20(12):99. doi: 10.1007/s11906-018-0899-2. [DOI] [PubMed] [Google Scholar]
  • 2.Ussavarungsi K., Thomas C.S., Burger C.D. Prevalence of metabolic syndrome in patients with pulmonary hypertension. Clin Respir J. 2017;11(6):721–726. doi: 10.1111/crj.12406. [DOI] [PubMed] [Google Scholar]
  • 3.Burger C.D., Foreman A.J., Miller D.P., Safford R.E., McGoon M.D., Badesch D.B. Comparison of body habitus in patients with pulmonary arterial hypertension enrolled in the Registry to Evaluate Early and Long-term PAH Disease Management with normative values from the National Health and Nutrition Examination Survey. Mayo Clin Proc. 2011;86(2):105–112. doi: 10.4065/mcp.2010.0394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Badesch D.B., Raskob G.E., Elliott C.G. Pulmonary arterial hypertension: baseline characteristics from the REVEAL Registry. Chest. 2010;137(2):376–387. doi: 10.1378/chest.09-1140. [DOI] [PubMed] [Google Scholar]
  • 5.Frank R.C., Min J., Abdelghany M. Obesity is associated with pulmonary hypertension and modifies outcomes. J Am Heart Assoc. 2020;9(5) doi: 10.1161/JAHA.119.014195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Robbins I.M., Newman J.H., Johnson R.F. Association of the metabolic syndrome with pulmonary venous hypertension. Chest. 2009;136(1):31–36. doi: 10.1378/chest.08-2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kenchaiah S., Evans J.C., Levy D. Obesity and the risk of heart failure. N Engl J Med. 2002;347(5):305–313. doi: 10.1056/NEJMoa020245. [DOI] [PubMed] [Google Scholar]
  • 8.Friedman S.E., Andrus B.W. Obesity and pulmonary hypertension: a review of pathophysiologic mechanisms. J Obes. 2012;2012:505274. doi: 10.1155/2012/505274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Assad T.R., Hemnes A.R., Larkin E.K. Clinical and biological insights into combined post- and pre-capillary pulmonary hypertension. J Am Coll Cardiol. 2016;68(23):2525–2536. doi: 10.1016/j.jacc.2016.09.942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Huston J.H., Maron B.A., French J. Association of mild echocardiographic pulmonary hypertension with mortality and right ventricular function. JAMA Cardiol. 2019;4(11):1112–1121. doi: 10.1001/jamacardio.2019.3345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Thayer T.E., Huang S., Levinson R.T. Unbiased phenome-wide association studies of red cell distribution width identifies key associations with pulmonary hypertension. Ann Am Thorac Soc. 2019;16(5):589–598. doi: 10.1513/AnnalsATS.201809-594OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Purcell S., Neale B., Todd-Brown K. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–575. doi: 10.1086/519795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Burgess S., Butterworth A., Thompson S.G. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658–665. doi: 10.1002/gepi.21758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Udler M.S., McCarthy M.I., Florez J.C., Mahajan A. Genetic risk scores for diabetes diagnosis and precision medicine. Endocr Rev. 2019;40(6):1500–1520. doi: 10.1210/er.2019-00088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Khera A.V., Chaffin M., Aragam K.G. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet. 2018;50(9):1219–1224. doi: 10.1038/s41588-018-0183-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gibson J., Russ T.C., Adams M.J. Assessing the presence of shared genetic architecture between Alzheimer's disease and major depressive disorder using genome-wide association data. Transl Psychiatry. 2017;7(4) doi: 10.1038/tp.2017.49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Walton E., Hibar D., Yilmaz Z. Exploration of shared genetic architecture between subcortical brain volumes and anorexia nervosa. Mol Neurobiol. 2019;56(7):5146–5156. doi: 10.1007/s12035-018-1439-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Garfield V., Fatemifar G., Dale C. Assessing potential shared genetic aetiology between body mass index and sleep duration in 142,209 individuals. Genet Epidemiol. 2019;43(2):207–214. doi: 10.1002/gepi.22174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Euesden J., Lewis C.M., O'Reilly P.F. PRSice: Polygenic Risk Score software. Bioinformatics. 2015;31(9):1466–1468. doi: 10.1093/bioinformatics/btu848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Yengo L., Sidorenko J., Kemper K.E. Meta-analysis of genome-wide association studies for height and body mass index in ∼700000 individuals of European ancestry. Hum Mol Genet. 2018;27(20):3641–3649. doi: 10.1093/hmg/ddy271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wells Q.S., Farber-Eger E., Crawford D.C. Extraction of echocardiographic data from the electronic medical record is a rapid and efficient method for study of cardiac structure and function. J Clin Bioinforma. 2014;4:12. doi: 10.1186/2043-9113-4-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22."2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension. The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS)." Nazzareno Galie, Marc Humbert, Jean-Luc Vachiery, Simon Gibbs, Irene Lang, Adam Torbicki, Gerald Simonneau, Andrew Peacock, Anton Vonk Noordegraaf, Maurice Beghetti, Ardeschir Ghofrani, Miguel Angel Gomez Sanchez, Georg Hansmann, Walter Klepetko, Patrizio Lancellotti, Marco Matucci, Theresa McDonagh, Luc A. Pierard, Pedro T. Trindade, Maurizio Zompatori and Marius Hoeper. Eur Respir J 2015; 46: 903-975. Eur Respir J. 2015;46(6):1855–1856. doi: 10.1183/13993003.51032-2015. [DOI] [PubMed] [Google Scholar]
  • 23.Price A.L., Patterson N.J., Plenge R.M., Weinblatt M.E., Shadick N.A., Reich D. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38(8):904–909. doi: 10.1038/ng1847. [DOI] [PubMed] [Google Scholar]
  • 24.Harrell F.E., Jr. Springer; New York; 2015. Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. [Google Scholar]
  • 25.Salem J.E., Shoemaker M.B., Bastarache L. Association of thyroid function genetic predictors with atrial fibrillation: a phenome-wide association study and inverse-variance weighted average meta-analysis. JAMA Cardiol. 2019;4(2):136–143. doi: 10.1001/jamacardio.2018.4615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yavorska O.O., Burgess S. MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. Int J Epidemiol. 2017;46(6):1734–1739. doi: 10.1093/ije/dyx034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Pugh M.E., Robbins I.M., Rice T.W., West J., Newman J.H., Hemnes A.R. Unrecognized glucose intolerance is common in pulmonary arterial hypertension. J Heart Lung Transplant. 2011;30(8):904–911. doi: 10.1016/j.healun.2011.02.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Zamanian R.T., Hansmann G., Snook S. Insulin resistance in pulmonary arterial hypertension. Eur Respir J. 2009;33(2):318–324. doi: 10.1183/09031936.00000508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Koop Anne-Marie C., Bossers Guido P.L., Ploegstra M.J. Metabolic remodeling in the pressure-loaded right ventricle: shifts in glucose and fatty acid metabolism—a systematic review and meta-analysis. J Am Heart Assoc. 2019;8(21) doi: 10.1161/JAHA.119.012086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Riehle C., Abel E.D. Insulin signaling and heart failure. Circ Res. 2016;118(7):1151–1169. doi: 10.1161/CIRCRESAHA.116.306206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Brittain E.L., Talati M., Fessel J.P. Fatty acid metabolic defects and right ventricular lipotoxicity in human pulmonary arterial hypertension. Circulation. 2016;133(20):1936–1944. doi: 10.1161/CIRCULATIONAHA.115.019351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Agarwal M., Agrawal S., Garg L., Lavie C.J. Relation between obesity and survival in patients hospitalized for pulmonary arterial hypertension (from a Nationwide Inpatient Sample Database 2003 to 2011) Am J Cardiol. 2017;120(3):489–493. doi: 10.1016/j.amjcard.2017.04.051. [DOI] [PubMed] [Google Scholar]
  • 33.Sharma A., Lavie C.J., Borer J.S. Meta-analysis of the relation of body mass index to all-cause and cardiovascular mortality and hospitalization in patients with chronic heart failure. Am J Cardiol. 2015;115(10):1428–1434. doi: 10.1016/j.amjcard.2015.02.024. [DOI] [PubMed] [Google Scholar]
  • 34.Weatherald J., Huertas A., Boucly A. Association between BMI and obesity with survival in pulmonary arterial hypertension. Chest. 2018;154(4):872–881. doi: 10.1016/j.chest.2018.05.006. [DOI] [PubMed] [Google Scholar]
  • 35.Sinning C., Harbaum L., Schrage B. Right ventricular index for risk stratification of patients with pulmonary arterial hypertension. Respiration. 2018;96(3):249–258. doi: 10.1159/000489231. [DOI] [PubMed] [Google Scholar]
  • 36.Henein M.Y., Gronlund C., Tossavainen E., Soderberg S., Gonzalez M., Lindqvist P. Right and left heart dysfunction predict mortality in pulmonary hypertension. Clin Physiol Funct Imaging. 2017;37(1):45–51. doi: 10.1111/cpf.12266. [DOI] [PubMed] [Google Scholar]
  • 37.Linderman G.C., Lu J., Lu Y. Association of body mass index with blood pressure among 1.7 million Chinese adults. JAMA Network Open. 2018;1(4) doi: 10.1001/jamanetworkopen.2018.1271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Lee M.-R., Lim Y.-H., Hong Y.-C. Causal association of body mass index with hypertension using a Mendelian randomization design. Medicine. 2018;97(30) doi: 10.1097/MD.0000000000011252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Larsson S.C., Back M., Rees J.M.B., Mason A.M., Burgess S. Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study. Eur Heart J. 2020;41(2):221–226. doi: 10.1093/eurheartj/ehz388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Reddy Y.N.V., Anantha-Narayanan M., Obokata M. Hemodynamic effects of weight loss in obesity. JACC Heart Failure. 2019;7(8):678. doi: 10.1016/j.jchf.2019.04.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wade K.H., Chiesa S.T., Hughes A.D. Assessing the causal role of body mass index on cardiovascular health in young adults: Mendelian randomization and recall-by-genotype analyses. Circulation. 2018;138(20):2187–2201. doi: 10.1161/CIRCULATIONAHA.117.033278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Dunlay S.M., Roger V.L., Redfield M.M. Epidemiology of heart failure with preserved ejection fraction. Nat Rev Cardiol. 2017;14(10):591–602. doi: 10.1038/nrcardio.2017.65. [DOI] [PubMed] [Google Scholar]
  • 43.Martin A.R., Gignoux C.R., Walters R.K. Human demographic history impacts genetic risk prediction across diverse populations. Am J Hum Genet. 2017;100(4):635–649. doi: 10.1016/j.ajhg.2017.03.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.LeVarge B.L., Pomerantsev E., Channick R.N. Reliance on end-expiratory wedge pressure leads to misclassification of pulmonary hypertension. Eur Respir J. 2014;44(2):425–434. doi: 10.1183/09031936.00209313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Jawad A., Tonelli A.R., Chatburn R.L., Wang X., Hatipoğlu U. Impact of intrathoracic pressure in the assessment of pulmonary hypertension in overweight patients. Ann Am Thorac Soc. 2017;14(12):1861–1863. doi: 10.1513/AnnalsATS.201704-331RL. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

e-Online Data
mmc1.pdf (701.5KB, pdf)

Articles from Chest are provided here courtesy of American College of Chest Physicians

RESOURCES