Smoking is associated with depletion of endothelial progenitor cells (EPCs) and may subsequently contribute to the development of vascular dysfunction. The aim of this study was to investigate the relation between circulating EPCs and pulmonary artery systolic pressure (PASP) as determined by flow cytometry and echocardiography in 174 patients (mean age 69 ± 9 years, 95 smokers) with established coronary artery disease. Smokers had significantly lower circulating log CD34/KDR + (0.86 ± 0.03 vs 0.96 ± 0.03 × 10 −3 /ml, p = 0.032) and log CD133/KDR + (0.68 ± 0.03 vs 0.82 ± 0.03 × 10 −3 /ml, p = 0.002) EPCs and a higher prevalence of elevated PASP >30 mm Hg (52% vs 30%, p = 0.001) than nonsmokers. Smokers with elevated PASP also had significantly lower circulating log CD34/KDR + (0.74 ± 0.04 vs 0.88 ± 0.06 × 10 −3 /ml, p <0.001) and log CD133/KDR + (0.61 ± 0.04 vs 0.78 ± 0.05 × 10 −3 /ml, p <0.001) EPCs, higher pulmonary vascular resistance, and larger right ventricular dimensions with impaired function (all p values <0.05). Log CD34/KDR + and log CD133/KDR + EPC counts were significantly and negatively correlated with PASP (r = −0.30, p <0.001, and r = −0.34, p <0.001, respectively) and pulmonary vascular resistance (r = −0.29, p = 0.002, and r = −0.18, p = 0.013, respectively). In conclusion, this study demonstrated that in patients with coronary artery disease, smoking was associated with a reduced number of EPCs and elevated PASP. This suggests that in smokers, depletion of circulating EPCs might be linked to the occurrence of pulmonary vascular dysfunction.
Pulmonary endothelial dysfunction plays an important role in the pathophysiology of pulmonary hypertension. It can be caused by direct injury to the endothelium, failure of repair, and compromise of barrier integrity. Previous experimental studies have shown that endothelial progenitor cells (EPCs) can be mobilized from the bone marrow into the circulation and contribute to the reendothelialization and neovascularization that occur after vascular injury. Depletion of circulating EPCs has been implicated in the development of vascular endothelial dysfunction. Recent studies have also demonstrated that dysfunction of EPCs may contribute to vascular remodeling in patients with pulmonary hypertension. Interestingly, nicotine has biphasic effects on EPCs: a low concentration has a positive influence on EPC number, proliferation, migration, and in vitro vasculogenesis, whereas a higher concentration is cytotoxic to EPCs. Depletion and dysfunction of EPCs have been observed in long-term smokers. We therefore hypothesized that cigarette smoking is associated with depletion of EPCs and may contribute to the development of elevated pulmonary artery systolic pressure (PASP) in patients with established coronary artery disease (CAD). Furthermore, we also aimed to define the clinical, laboratory, and echocardiographic predictors of elevated PASP in patients with CAD.
Methods
This was a cross-sectional observational study on the relation between circulating EPCs and PASP in smokers with CAD. Consecutive patients with CAD documented by coronary angiography were recruited from the outpatient clinic. All patients had ≥50% stenosis of ≥1 of the major coronary arteries and had received consistent medication for ≥1 year before enrollment. Patients with dilated cardiomyopathy, significant valvular heart disease, chronic atrial fibrillation, recent myocardial infarctions, unstable angina, coronary revascularization or strokes within the past 6 months, New York Heart Association class III or IV heart failure, histories of any chronic lung disease, peripheral vascular disease, autoimmune or connective tissue disease, or renal or liver dysfunction and those who decreased to participate were excluded. To determine the range of circulating EPCs in normal subjects, 65 age- and gender-matched subjects who had no known histories of cardiovascular diseases and were not taking any cardiovascular medications, including statins, were enrolled as controls. The study was approved by the institutional review board, and all subjects gave written informed consent.
Baseline demographic data, cardiovascular risk factors, and cardiovascular medications were documented in all subjects. Cardiovascular risk factors, including cigarette smoking, diabetes mellitus, hypercholesterolemia, hypertension, body mass index, and family history of CAD in first-degree relatives aged <55 years were also assessed as previously described. Hypertension was defined as systolic or diastolic blood pressure at rest ≥140 or 90 mm Hg, respectively, at 2 different clinic visits or the prescription of antihypertensive medication. Diabetes mellitus was defined as serum fasting glucose ≥7.0 mmol/L or the use of hypoglycemic agents. Hypercholesterolemia was defined as fasting total plasma cholesterol ≥4.9 mmol/L. Smoking status was recorded as smoker (current or former) or nonsmoker. Fasting blood samples were obtained from all subjects to determine serum creatinine, glucose, and lipid levels, including total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglyceride, apolipoprotein, apolipoprotein B, and lipoprotein(a).
All patients underwent detailed 2-dimensional, M-mode, Doppler flow, and tissue Doppler imaging echocardiography in the left lateral decubitus position using a digital ultrasound system (Vivid 7; GE Medical Systems, Milwaukee, Wisconsin) using a 3- to 5-MHz transducer. Standard 2-dimensional and M-mode measurements were obtained according to the guidelines of the American Society of Echocardiography. The left ventricular (LV) ejection fraction was measured using the modified biplane Simpson’s method from apical 4- and 2-chamber views. The level of tricuspid annular plane systolic excursion was obtained by drawing a straight line through the lateral tricuspid valve annulus under M mode as an index of right ventricular (RV) function.
Mitral inflow E velocity and its deceleration time and mitral inflow A velocity were measured using pulse-wave Doppler with the sample volume placed at the tips of the mitral valve from the apical 4-chamber view. Continuous-wave Doppler was used to determine the maximal tricuspid regurgitation velocity. RV systolic pressure was estimated from tricuspid regurgitation velocity by modification of the Bernoulli formula. PASP was calculated by adding the sum of RV systolic pressure and mean right atrial pressure measured using continuous-wave Doppler echocardiography. Elevated pulmonary artery pressure was defined as PASP >30 mm Hg. Pulmonary vascular resistance (PVR) was calculated using the validated formula tricuspid regurgitation velocity/velocity-time integral of RV outflow tract × 10 + 0.16. A pulsed tissue Doppler imaging spectrum was also obtained from the apical 4-chamber view. Angle, depth, and pulse repetitive frequency were adjusted to the possible highest frame rate (usually ≥150 frames/s). Mitral annular and tricuspid annular peak velocities during systole (Vs) and early diastole (Ve) were obtained using pulsed tissue Doppler spectra with the sample volume placed at the septal and lateral mitral annulus and at the tricuspid annulus of the RV free wall, respectively. The ratio of mitral inflow E velocity to septal mitral annulus early diastolic (E′) velocity (E/E′) was calculated as an index of LV filling pressure. Diastolic function was categorized as normal filling pattern (mitral inflow E/A ratio >1.0, deceleration time >220 and <280 ms, and E/E′ sep <8) or abnormal relaxation filling pattern (mitral inflow E/A ratio <1.0, deceleration time >280 ms, and E/E′ sep >12).
All echocardiographic examinations were performed by 2 experienced operators (W-SY or G-HY), and all digital images were stored on optical discs for subsequent off-line analysis by another operator (MW), who was blinded to patients’ clinical details. At least 3 consecutive beats were analyzed off-line using a computer workstation (EchoPAC; GE Medical Systems). Interobserver variability testing confirmed correlation coefficients of 0.78 to 0.83 (p <0.01) for echocardiographic measurement.
Circulating EPCs were defined by their expression of surface markers CD34/KDR + and CD133/KDR + , and their numbers were measured using fluorescence-activated cell analysis of a peripheral blood sample, as described previously. In brief, 100 μl of peripheral blood was incubated with a phycoerythrin-conjugated monoclonal antibody against human KDR (Sigma, St. Louis, Missouri), followed by fluorescein isothiocyanate–conjugated CD34 and CD133 antibodies (Beckman Coulter, Fullerton, California). Fluorescein isothiocyanate–labeled antihuman CD45 antibody was used for differential gating during flow analysis. Fluorescein isothiocyanate–labeled immunoglobulin G1a (Beckman Coulter) and phycoerythrin-labeled immunoglobulin G2b (Becton Dickinson, Franklin Lakes, New Jersey) served as the isotypic control for color compensation. Analysis was performed with an automated fluorescence-activated cell counter (Elite; Beckman Coulter) in which 1 million events were counted. Intraobserver variability testing confirmed an intraclass correlation coefficient of 0.9 (p <0.001).
Continuous variables are expressed as mean ± SEM. Categorical data are presented as frequencies and percentages. Because the distribution patterns of the number of EPCs were highly skewed, these variables were log transformed to normalize their distribution before analysis. Statistical comparisons were performed using Student’s t test or the chi-square test, as appropriate. Comparisons of variables among different groups were performed with 1-way analysis of variance with post hoc Bonferroni’s correction for comparison among multiple groups. The relation of EPCs to PASP and to other echocardiographic parameters was analyzed using a linear regression and expressed as Pearson’s bivariate correlation coefficient. To have ≥80% power to detect a 10% difference in EPCs between smokers and nonsmokers with a 5% maximum false-positive error rate, we would need ≥75 patients from each group. All statistical analyses were performed using SPSS version 16.0 (SPSS, Inc., Chicago, Illinois). A p value <0.05 was considered statistically significant.
Results
A total of 174 patients with CAD were enrolled (151 men and 23 women, mean age 69 ± 1 years, range 36 to 88). Ninety-five patients were current or former smokers and 79 were nonsmokers. The clinical characteristics of the study population are listed in Table 1 . There were no significant differences in age, prevalence of male gender, hypertension, diabetes, hypercholesterolemia, previous myocardial infarction or coronary revascularization, systolic and diastolic blood pressure, body mass index, and medications between smokers and nonsmokers (all p values >0.05; Table 1 ). There were no significant differences in serum fasting blood glucose, glycosylated hemoglobin, triglycerides, total cholesterol, low-density lipoprotein cholesterol, apolipoprotein B, lipoprotein(a), apolipoprotein A-I, and creatinine levels between smokers and nonsmokers (all p values >0.05; Table 1 ). Smokers had significantly lower serum high-density lipoprotein cholesterol levels and circulating CD34/KDR + and CD133/KDR + EPC levels (all p values <0.05; Table 1 ).
| Variable | Smokers With CAD (n = 97) | Nonsmokers With CAD (n = 67) | p Value † | Controls Without CAD and Smokers (n = 65) | p Value ‡ | p Value § |
|---|---|---|---|---|---|---|
| Age (years) | 67 ± 1 | 67 ± 1 | 0.87 | 66 ± 1 | 0.41 | 0.52 |
| Men | 81 (83%) | 51 (77%) | 0.33 | 50 (78%) | 0.30 | 0.69 |
| Previous myocardial infarction | 51 (53%) | 26 (40%) | 0.062 | |||
| Previous revascularization | 31 (18%) | 10 (14%) | 0.79 | |||
| Hypertension | 64 (66%) | 40 (60%) | 0.52 | 10 (15%) | <0.001 | <0.001 |
| Diabetes mellitus | 36 (37%) | 25 (38%) | 0.87 | 8 (12%) | <0.001 | <0.001 |
| Hypercholesterolemia ⁎ | 65 (68%) | 40 (60%) | 0.34 | 25 (37%) | <0.001 | 0.001 |
| BMI (kg/m 2 ) | 25.3 ± 0.3 | 24.8 ± 0.4 | 0.34 | 23.7 ± 0.39 | 0.001 | 0.086 |
| Blood pressure (mm Hg) | ||||||
| Systolic | 143.5 ± 1.9 | 142.9 ± 2.1 | 0.86 | 118.2 ± 2.28 | <0.001 | <0.001 |
| Diastolic | 81.8 ± 0.8 | 83.1 ± 1.0 | 0.31 | 72.0 ± 1.1 | <0.001 | <0.001 |
| Biochemical analysis | ||||||
| Fasting blood glucose (mg/dl) | 110.3 ± 1.8 | 108.5 ± 4.9 | 0.77 | 88.7 ± 1.1 | <0.001 | <0.001 |
| Hb A1c (%) | 6.9 ± 0.2 | 6.6 ± 0.2 | 0.14 | 6.0 ± 0.1 | <0.001 | 0.003 |
| Triglycerides (mg/dl) | 138.0 ± 12.5 | 123.7 ± 8.9 | 0.43 | 114.8 ± 8.0 | 0.19 | 0.42 |
| Total cholesterol (mg/dl) | 162.2 ± 2.7 | 157.2 ± 2.7 | 0.25 | 199.3 ± 3.5 | <0.001 | <0.001 |
| LDL cholesterol (mg/dl) | 91.3 ± 2.0 | 86.2 ± 2.7 | 0.14 | 113.5 ± 3.1 | <0.001 | <0.001 |
| HDL cholesterol (mg/dl) | 45.2 ± 1.2 | 49.5 ± 1.6 | 0.032 | 62.8 ± 2.3 | <0.001 | <0.001 |
| Apo A-I (g/L) | 1.22 ± 0.02 | 1.29 ± 0.03 | 0.092 | 1.38 ± 0.03 | <0.001 | 0.030 |
| Apo B (g/L) | 0.74 ± 0.02 | 0.71 ± 0.02 | 0.182 | 0.91 ± 0.02 | <0.001 | <0.001 |
| Lp(a) (mg/L) | 283.7 ± 24.9 | 267.5 ± 22.0 | 0.35 | 282.4 ± 23.2 | 0.980 | 0.33 |
| Creatinine (μmol/L) | 91.2 ± 1.9 | 93.2 ± 2.2 | 0.63 | 72.6 ± 1.9 | <0.001 | <0.001 |
| Medications | ||||||
| β blockers | 72 (74%) | 55 (82%) | 0.15 | 4 (7%) | <0.001 | <0.001 |
| Calcium channel blockers | 26 (27%) | 13 (19%) | 0.28 | 6 (9%) | <0.001 | 0.166 |
| ACE inhibitors/ARBs | 76 (79%) | 47 (70%) | 0.37 | 0 (0%) | <0.001 | <0.001 |
| Nitrates | 48 (50%) | 36 (54%) | 0.64 | 0 (0%) | <0.001 | <0.001 |
| Aspirin | 63 (95%) | 63 (93%) | 0.73 | 3 (5%) | <0.001 | <0.001 |
| Statins | 93 (96%) | 62 (93%) | 0.31 | 2 (3%) | <0.001 | <0.001 |
| Log CD34/KDR + EPCs (×10 −3 /ml) | 0.85 ± 0.04 | 0.97 ± 0.03 | 0.027 | 1.31 ± 0.42 | <0.001 | <0.001 |
| Log CD133/KDR + EPCs (10 −3 /ml) | 0.67 ± 0.04 | 0.79 ± 0.03 | 0.003 | 0.92 ± 0.03 | <0.001 | 0.016 |
⁎ Fasting total plasma cholesterol ≥4.9 mmol/L.
† Smokers with CAD versus nonsmokers with CAD.
‡ Smokers with CAD versus controls without CAD and smokers.
§ Nonsmokers with CAD versus controls without CAD and smokers.
Compared to controls, smokers or smokers with CAD had a significantly higher prevalence of cardiovascular risk factors, including hypertension, diabetes, hyperlipidemia, and elevated blood pressure, body mass index, fasting blood glucose, and glycosylated hemoglobin (all p values <0.05; Table 1 ). Furthermore, controls had higher levels of high-density lipoprotein cholesterol, apolipoprotein B and A-I, and circulating CD34/KDR + and CD133/KDR + EPCs compared to patients with CAD (all p values <0.05; Table 1 ).
As listed in Table 2 , 2-dimensional and M-mode echocardiography showed that smokers had significantly larger LV, left atrial, and RV dimensions and lower tricuspid annular plane systolic excursion compared to nonsmokers. There were nonetheless no significant differences in the LV ejection fraction and right atrial diameter between the 2 groups (all p values >0.05; Table 2 ). Doppler echocardiography revealed that the prevalence of elevated PASP and mean PASP were significantly higher in smokers than nonsmokers (all p values <0.05; Table 2 ). Nevertheless, there were no significant differences in the LV diastolic index, the prevalence of LV diastolic dysfunction, mitral regurgitation, or PVR (all p values >0.05; Table 2 ). Tissue Doppler imaging demonstrated significantly lower Vs and Ve of the RV free wall in smokers (all p values <0.05; Table 2 ). There were no significant differences in Vs and Ve of the septal and lateral mitral annulus and the E/E′ ratio between the 2 groups (all p values >0.05; Table 2 ).
| Variable | Smokers (n = 95) | Nonsmokers (n = 79) | p Value |
|---|---|---|---|
| LV end-diastolic diameter (mm) | 49.3 ± 0.7 | 47.1 ± 0.7 | 0.033 |
| LV end-systolic diameter (mm) | 34.1 ± 0.8 | 31.3 ± 0.8 | 0.026 |
| Left atrial dimension (mm) | 38.9 ± 0.5 | 37.3 ± 0.5 | 0.038 |
| LV ejection fraction (%) | 570.0 ± 1.4 | 61.6 ± 1.3 | 0.081 |
| RV end-diastolic diameter (mm) | 62.2 ± 0.7 | 60.5 ± 0.8 | 0.037 |
| RV end-systolic diameter (mm) | 45.8 ± 0.8 | 44.8 ± 0.7 | 0.042 |
| Right atrial dimension (mm) | 32.3 ± 2.0 | 360.0 ± 2.0 | 0.082 |
| TAPSE (mm) | 2.02 ± 0.04 | 1.90 ± 0.04 | 0.037 |
| Mitral inflow E (m/s) | 0.67 ± 0.02 | 0.68 ± 0.02 | 0.77 |
| Mitral inflow A (m/s) | 0.83 ± 0.02 | 0.85 ± 0.02 | 0.62 |
| E/A | 0.86 ± 0.05 | 0.84 ± 0.03 | 0.66 |
| Deceleration time (ms) | 229 ± 7 | 229 ± 6 | 0.99 |
| LV septal Ve (E′) (cm/s) | 5.3 ± 0.1 | 5.7 ± 0.2 | 0.59 |
| LV lateral wall Ve (cm/s) | 70.4 ± 0.3 | 7.6 ± 0.2 | 0.78 |
| RV free wall Ve (cm/s) | 9.2 ± 0.3 | 9.8 ± 0.4 | 0.041 |
| E/E′ | 13.8 ± 0.9 | 12.1 ± 0.5 | 0.10 |
| LV septal Vs (cm/s) | 5.8 ± 0.2 | 6.3 ± 0.2 | 0.13 |
| LV lateral wall Vs (cm/s) | 6.9 ± 0.3 | 7.7 ± 0.3 | 0.10 |
| RV free wall Vs (cm/s) | 10.9 ± 0.2 | 11.7 ± 0.2 | 0.045 |
| Abnormal diastolic pattern | 48 (51%) | 34 (43%) | 0.37 |
| PASP (mm Hg) | 33 ± 1.0 | 30 ± 0.8 | 0.037 |
| PHT | 50 (52%) | 24 (30%) | 0.001 |
| PVR (Wood units) | 1.26 ± 0.06 | 1.25 ± 0.06 | 0.91 |
| Mitral regurgitation | 70 (74%) | 72 (83%) | 0.11 |
As listed in Table 3 , there were no significant differences in age, the proportion of men, the prevalence of hypertension, diabetes, hypercholesterolemia, myocardial infarction and coronary revascularization, systolic and diastolic blood pressure, body mass index, and medications between smokers with or without elevated PASP (all p values >0.05). There were also no significant differences in serum fasting blood glucose, glycosylated hemoglobin, lipid profile, and creatinine levels (all p values >0.05; Table 3 ). However, smokers with elevated PASP had significantly lower circulating CD34/KDR + and CD133/KDR + EPC levels than those without elevated PASP (all p values <0.001; Table 3 ).
| Variable | With Elevated PASP (n = 50) | Without Elevated PASP (n = 45) | p Value |
|---|---|---|---|
| Age (years) | 69 ± 1 | 67 ± 1 | 0.23 |
| Men | 48 (95%) | 44 (98%) | 0.63 |
| Previous myocardial infarction | 28 (56%) | 26 (53%) | 0.84 |
| Previous revascularization | 12 (24%) | 5 (12%) | 0.25 |
| Hypertension | 35 (70%) | 24 (53%) | 0.11 |
| Diabetes mellitus | 17 (34%) | 16 (36%) | 1.00 |
| Hypercholesterolemia ⁎ | 33 (66%) | 62 (28%) | 0.69 |
| BMI (kg/m 2 ) | 24.8 ± 0.3 | 25.7 ± 0.5 | 0.17 |
| Blood pressure (mm Hg) | |||
| Systolic | 144.5 ± 2.5 | 140.2 ± 2.8 | 0.27 |
| Diastolic | 80.9 ± 1.0 | 81.9 ± 1.4 | 0.57 |
| Biochemical analysis | |||
| Fasting blood glucose (mg/dl) | 106.9 ± 4.1 | 113.9 ± 6.5 | 0.36 |
| Hb A1c (%) | 6.8 ± 0.2 | 7.1 ± 0.3 | 0.38 |
| Triglycerides (mg/dl) | 134.4 ± 17.8 | 141.5 ± 19.6 | 0.80 |
| Total cholesterol (mg/dl) | 166.5 ± 4.3 | 157.2 ± 3.1 | 0.071 |
| LDL cholesterol (mg/dl) | 94.4 ± 3.1 | 87.0 ± 2.7 | 0.088 |
| HDL cholesterol (mg/dl) | 44.1 ± 1.2 | 44.9 ± 2.0 | 0.82 |
| Apo A-I (g/L) | 1.22 ± 0.03 | 1.20 ± 0.20 | 0.66 |
| Apo B (g/L) | 0.72 ± 0.13 | 0.78 ± 0.16 | 0.063 |
| Lp(a) (mg/L) | 263 ± 38 | 257 ± 40 | 0.92 |
| Creatinine (μmol/L) | 97.5 ± 3.7 | 97.2 ± 3.9 | 0.95 |
| Medications | |||
| β blockers | 36 (73%) | 33 (73%) | 1.00 |
| Calcium channel blockers | 13 (27%) | 18 (8%) | 0.35 |
| ACE inhibitors/ARBs | 40 (80%) | 34 (76%) | 0.64 |
| Nitrates | 23 (46%) | 25 (55%) | 0.43 |
| Aspirin | 47 (95%) | 44 (98%) | 0.63 |
| Statins | 48 (97%) | 42 (93%) | 0.65 |
| Log CD34/KDR + EPCs (×10 −3 /ml) | 0.74 ± 0.04 | 0.88 ± 0.06 | <0.001 |
| Log CD133/KDR + EPCs (×10 −3 /ml) | 0.61 ± 0.04 | 0.78 ± 0.05 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
