Endothelial Function and Soluble Endoglin in Smokers With Heart Failure

Abstract

Background:

Although cigarette smoking is a risk factor for heart failure (HF), smokers with HF have lower mortality rates during/following hospitalization compared to nonsmokers. We examined vascular endothelial function in chronic smokers and nonsmokers with HF as it relates to this smoker's paradox.

Hypothesis:

Smokers with HF will have attenuated endothelial dysfunction compared to non‐smokers with HF.

Methods:

Brachial artery flow‐mediated dilation (FMD), a measure of conduit vessel endothelial function, was measured in 33 smoking and nonsmoking patients with HF vs controls. In addition, soluble endoglin (sEng), a circulating mediator of endothelial function, was measured in a separate group of 36 smoking and nonsmoking patients with HF vs controls.

Results:

FMD was significantly lower in smokers without HF compared to the nonsmokers without HF (P < 0.05). FMD was significantly higher in smokers with HF vs nonsmokers with HF (P < 0.05) and did not differ from values seen in nonsmokers without HF (P > 0.05). There were no differences in sEng between smokers and nonsmokers without HF (P > 0.05). sEng was lower in smokers with HF vs nonsmokers with HF (P < 0.05) and did not differ from values seen in nonsmokers without HF (P > 0.05).

Conclusions:

Smokers with HF had higher brachial FMD and lower sEng than nonsmokers with HF, and values were comparable to nonsmokers without HF. These findings offer novel insight into the smoker's paradox and suggest that improved short‐term outcome in patients hospitalized with HF may in part be mediated by preservation of vascular endothelial function in this setting. © 2011 Wiley Periodicals, Inc.

The authors have no funding, financial relationships, or conflicts of interest to disclose.

Introduction

Impaired endothelial function is highly prevalent if not ubiquitous in patients with heart failure (HF).1 Endothelial dysfunction in the setting of HF is predictive of need for hospitalization,2 cardiac transplantation,2 response to therapy,3 and mortality.4, 5 Cigarette smoking also causes pervasive vascular endothelial dysfunction.6 Although cigarette smoking is an established risk factor for the development of HF, smokers with HF have paradoxically lower short‐term mortality rates (in‐hospital and early postdischarge) compared to nonsmokers.7 In isolated clinical settings, smoking may have paradoxically favorable effects on endothelial biology and outcome.8 Whether there are differences in systemic vascular endothelial function in smokers and nonsmokers with HF has yet to be investigated.

The purpose of this study was to test the hypothesis that endothelial dysfunction would be attenuated in smokers with HF compared to nonsmokers with HF in vivo, offering novel insight into the smoker's paradox. In our first approach, we examined endothelial‐dependent vasodilation of the brachial artery in smokers and nonsmokers with HF compared to controls. In our second approach, we quantified circulating levels of soluble endoglin (sEng) (a regulator of endothelial vasomotor tone) in smokers and nonsmokers with HF compared to controls.

Methods

Patient Population

We examined 33 patients with heart failure (New York Heart Association class II and III and left ventricular ejection fraction ≤30%) who were referred for right and left heart catheterization at Tufts Medical Center, Boston, Massachusetts. Control subjects consisted of 43 volunteers without a history of HF or any signs/symptoms suggestive of current HF. All subjects were required to be between 21 and 80 years of age. Patients under 18 years of age and those presenting with an acute coronary syndrome, pregnancy, active or remote cancer, renal failure (estimated glomerular filtration rate ≤30), liver transaminases >2 times the upper limit of normal, non‐sinus rhythm, or perceived interference with standard clinical care were excluded. Left ventricular ejection fraction was assessed by echocardiography or ventriculography using standard techniques. The institutional review board of Tufts Medical Center approved this study and all patients provided written informed consent.

Approach I

In the first study, 33 subjects underwent assessment of brachial artery flow‐mediated dilation (FMD) using high‐resolution ultrasonography following conventional laboratory protocol. The brachial artery was longitudinally imaged 2 cm above the antecubital fossa using a 10‐mHz linear array vascular ultrasound transducer. Following baseline diameter measurement with ultrasonic calipers (obtained during end‐diastole), reactive hyperemia was induced by inflation of a blood pressure cuff around the upper arm to a suprasystolic pressure for 5 minutes. Sixty seconds following release of the occlusion cuff, brachial diameter was once again measured as previously mentioned. Responses were calculated as percentage change in brachial artery diameter from baseline (FMD).

Approach II

In the second study, 36 patients had serum samples obtained for biomarker assay. HF patients had blood sampled at the time of arterial or venous sheath insertion for diagnostic catheterization. Control subjects had blood taken during a one‐time lab draw in our clinical research center. Human sEng, B‐type natriuretic peptide (BNP), and atrial natriuretic peptide (ANP), were measured in each serum sample in duplicate using commercially available quantitative sandwich enzyme immunoassay kits (Soluble Endoglin; R&D Systems, Minneapolis, MN/BNP and ANP; Phoenix Pharmaceuticals, Burlingame, CA) according to the manufacturers' instructions. The intra‐assay and interassay coefficient of variation for each protein assay were: sEng, 3% and 6%, respectively; BNP, 8% and 4%, respectively; and ANP, 11% and 5%, respectively.

Statistical Analysis

All data are reported as mean ± standard error of mean. A priori significance was set at P < 0.05. Analysis of variance was used to assess differences in continuous variables between groups with Tukey testing for post hoc comparisons. Analysis of covariance (ANCOVA) was used to statistically adjust for the effect of potential confounders, and χ 2 tests were used to compare categorical variables. Data analysis was carried out using SPSS software (SPSS, Inc., Chicago, IL).

Results

Approach I

Patient characteristics are presented in Table 1. Groups were of similar age, body surface area, and gender and had similar cardiovascular (CV) risk factor profiles. Diuretic use was significantly higher (P < 0.05) in patients with HF (SM⁺ and SM⁻) versus controls (SM⁺ and SM⁻). FMD was significantly lower in nonsmokers with HF compared to nonsmoking controls (P < 0.05). FMD was significantly lower in smokers without HF compared the nonsmoking controls (P < 0.05). FMD was significantly higher in smokers with HF vs nonsmoking controls (P < 0.05). FMD did not differ between smokers with HF and nonsmoking controls (P > 0.05). Adjusting for group differences in diuretic use had no effect on differences in FMD (P < 0.05).

Table 1.

Descriptive Characteristics for Smokers and Nonsmokers With and Without Heart Failure

Variable	SM+HF+, n = 7	SM−HF+, n = 8	SM+HF−, n = 13	SM−HF−, n = 12
Age, y	59 ± 4	54 ± 3	56 ± 2	56 ± 4
Body surface area, m2	1.99 ± 0.07	2.04 ± 0.07	2.03 ± 0.06	1.99 ± 0.07
Female, %	0	50	15	25
Hypertension, %	57	75	69	60
Dyslipidemia, %	100	62	85	60
Diabetes mellitus, %	14	38	23	10
CAD, %	57	38	77	50
NYHA class (0–4)	2	2	0a,b	0a,b
Ischemic cardiomyopathy, %	57	38	—	—
Ejection fraction, %	18 ± 3	25 ± 3	57 ± 2a,b	58 ± 3a,b
Medications, %
Statin	57	50	69	50
ASA	57	38	62	50
β‐Blocker	86	38	54	40
Ca channel blocker	0	0	13	30
ACE inhibitor	29	75	54	30
Diuretic	57	75	8a,b	0a,b
Baseline diameter, mm	3.6 ± 0.2	3.8 ± 0.1	4.1 ± 0.2	4.0 ± 0.2
Brachial FMD, %c	14.0 ± 2.4	7.2 ± 2.5a	7.2 ± 1.8a	15.2 ± 1.9b,d

Abbreviations: ACE, angiotensin‐converting enzyme; ASA, acetylsalicylic acid; CAD, coronary artery disease; FMD, flow‐mediated dilation; NYHA, New York Heart Association.

^aSignificantly different than SM⁺/HF⁺.

^bSignificantly different than SM⁻/HF⁺.

^cAdjusted for group differences in diuretic use.

^dSignificantly different than SM⁺/HF⁻.

Approach II

Patient characteristics are presented in Table 2. Groups were of similar age, body surface area, and gender and had similar CV risk factor profiles. Diuretic use was higher (P < 0.05) in patients with HF (SM⁺ and SM⁻) vs controls (SM⁺ and SM⁻). Left ventricular end‐diastolic pressure (LVEDP) was higher in nonsmokers with HF compared to control groups (P < 0.05). Nonsmokers with HF had significantly higher sEng and BNP compared to all other groups (P < 0.05). sEng and BNP were similar (P > 0.05) between smokers with HF and controls (SM⁺ and SM⁻). Smokers with HF maintained lower values than nonsmokers with HF after adjusting for BNP, LVEDP, and diuretic use with ANCOVA (P < 0.05).

Table 2.

Descriptive Characteristics for Smokers and Nonsmokers With and Without Heart Failure

Variable	SM+HF+, n = 8	SM−HF+, n = 10	SM+HF−, n = 7	SM−HF−, n = 11
Age, y	57 ± 4	59 ± 5	62 ± 3	59 ± 3
Body surface area, m2	1.84 ± 0.09	1.94 ± 0.07	1.98 ± 0.07	1.92 ± 0.07
Female, %	75	60	43	73
Hypertension, %	75	50	100	91
Dyslipidemia, %	75	50	100	45
Diabetes mellitus, %	25	30	43	18
CAD, %	25	40	43	18
NYHA class (0–4)	2	3	0a,b	0a,b
Ischemic cardiomyopathy, %	25	40	—	—
LV ejection fraction, %	26 ± 6	30 ± 6	57 ± 2a,b	59 ± 1a,b
LV end‐diastolic pressure, mm Hg	15 ± 6	21 ± 7	11 ± 3b	10 ± 3b
Medications, %
Statin	75	50	100	45
ASA	62	50	100	73
β‐Blocker	50	60	100	45
Ca channel blocker	12	40	14	9
ACE inhibitor	38	30	43	36
Diuretic	75	90	14a,b	18a,b
ANP, ng/mL	50 ± 17	55 ± 12	22 ± 10	18 ± 5a
BNP, ng/mL	10 ± 1	20 ± 3a	9 ± 2b	8 ± 2b
sEng, pg/mL	4110 ± 217	5305 ± 274a	3554 ± 209a,b	3372 ± 152a,b

Abbreviations: ACE, angiotensin‐converting enzyme; ANP, atrial natriuretic peptide; ASA, acetylsalicylic acid; BNP, B‐type natriuretic peptide; CAD, coronary artery disease; FMD, flow‐mediated dilation; LV, left ventricular; NYHA, New York Heart Association; sEng, soluble endoglin.

^aSignificantly different than SM⁺HF⁺.

^bSignificantly different than SM⁻/HF⁺.

Discussion

Smokers without HF had significantly lower brachial FMD compared to their nonsmoking counterparts without HF, underscoring established findings of the detrimental effect of smoking on the vascular endothelium. The novel findings of the present study were that smokers with HF had higher brachial FMD and lower sEng compared to nonsmokers with HF. Interestingly, brachial FMD and sEng levels in smokers with HF were comparable to nonsmokers without HF. Taken together, these findings suggest that compared to nonsmokers with HF, smokers with HF have preserved endothelial function.

Smoking increases the risk of HF. Despite this, smokers with HF have improved short‐term prognosis during/following hospitalization, a phenomenon referred to as the smoker's paradox.2 This paradox is also apparent in other disease states, as smokers with preeclampsia,8, 9 stroke,10, 11 and AMI12, 13 have more favorable clinical outcomes than nonsmokers. These findings have been mostly attributed to differences between smokers and nonsmokers on presentation, as smokers tend to be younger with a lower prevalence of comorbidities. However, even after adjustment for baseline differences in age and clinical demographic features, the survival advantage witnessed in smokers during these isolated events prevails.

In the present study, we noted that nonsmokers with HF had lower brachial FMD than nonsmokers without HF, recapitulating previous reports of a well‐noted endothelial dysfunction in HF.1, 4, 5, 14 Our results also support numerous studies documenting that smoking reduces endothelial function,6 as smokers without HF had significantly lower brachial FMD as nonsmokers without HF. The novel finding of the present study was that smokers with HF had higher brachial FMD compared to nonsmokers with HF. Moreover, FMD values in these patients were similar to values seen in nonsmoking control patients. Thus, our findings suggest that the improved short‐term prognosis previously reported in smokers with HF may be related to preservation of vascular endothelial integrity during and/or following hospitalization.

Smokers with HF had lower circulating sEng than nonsmokers with HF, offering novel mechanistic insight into the aforementioned findings of preserved endothelial function in smokers with HF. Endoglin is a TGF‐β coreceptor highly expressed in endothelial cells.15 Endoglin has been shown to regulate the expression and activity of endothelial NO synthase,16 reducing NO production17 and making it an important modulator of endothelial vasomotor tone.18 sEng is increased in the setting of HF.19 Excess endoglin has been shown to cause endothelial dysfunction by inhibiting TGF‐β–mediated vasorelaxation and increasing vascular resistance.20 In other pathologies, smoking is associated with lower circulating levels of sEng.8 Cigarette smoke extract and nicotine have been shown to reduce release of sEng,21, 22, 23 and nicotine restores endothelial dysfunction (ie, endothelial cell migration and tube formation in response to vascular injury) caused by excess sEng in vitro.24 Cigarette smoke also increases heme oxygenase expression in endothelial cells,25, 26 the key enzyme involved in carbon monoxide generation (a potent vasodilator).27 Heme oxygenase and carbon monoxide are potent negative regulators of sEng release.28 Thus, it is possible that smoking and its active constituents may be preconditioning the vasculature to subsequent insults during pathological triggers such as HF. Preservation of endothelial integrity with progression of HF may further portend a vasculo‐protective effect, improving short‐term outcome.

Conclusion

HF appears to modulate the relationship between smoking exposure and endothelial biology. Smokers with HF have preserved endothelial function and lower sEng compared to nonsmokers with HF. This may exert a cardio‐protective effect during and following hospitalization in HF, offering novel insight into previous reports of improved short‐term survival in smokers with HF.