Circulating Progenitor Cells and Coronary Collaterals in Chronic Total Occlusion

Abstract

Background:

The role of circulating progenitor cells (CPC) in collateral formation that occurs in the presence of chronic total occlusions (CTO) of a coronary artery is not well established. In stable patients with a CTO, we investigated whether CPC levels are associated with (a) collateral development and (b) ischemic burden, as measured by circulating high sensitivity troponin-I (hsTn-I) levels.

Methods:

CPCs were enumerated by flow cytometry as CD45^med+ blood mononuclear cells expressing CD34 and both CD34 and CD133 epitopes. The association between CPC counts and both Rentrop collateral grade (0, 1, 2, or 3) and hsTn-I levels were evaluated using multivariate regression analysis, after adjusting for demographic and clinical characteristics.

Results:

In 89 patients (age 65.5, 72% male, 27% Black), a higher CPC count was positively associated with a higher Rentrop collateral grade; [CD34+ adjusted odds ratio (OR) 1.49 95% confidence interval (CI) (0.95, 2.34) P = 0.082] and [CD34+/CD133+ OR 1.57 95% CI (1.05, 2.36) P = 0.028]. Every doubling of CPC counts was also associated with lower hsTn-I levels [CD34+ β −0.35 95% CI (−0.49, −0.15) P = 0.002] and [CD34+/CD133+ β −0.27 95% CI (−0.43, −0.08) P = 0.009] after adjustment.

Conclusion:

Individuals with higher CPC counts have greater collateral development and lower ischemic burden in the presence of a CTO.

1. Introduction

Progenitor and stem cells in the circulation are largely derived from the bone marrow and contribute to repair of the vasculature and myocardium after injury [1–4]. Circulating progenitor cells (CPCs) that express the cluster of differentiation (CD) 34 epitope have the potential to differentiate into several distinct lineages that contribute to endogenous repair capacity such as hematopoietic cells (identified by their expression of the CD34 epitope on CD45^med cells), and nonhematopoietic mesenchymal progenitor cells, which lack CD45 expression [5–8]. CD133 is a 5-transmembrance antigen that is lost during maturation and therefore identifies an early CPC-enriched subpopulation [9,10].

In the setting of injury, progenitor cells migrate from the bone marrow into the circulation, to the culprit tissue and contribute to repair and regeneration [11–13]. Previous studies have demonstrated that patients have a variable ability to mobilize progenitor cells [14,15]. Moreover, low CPC counts are an independent indicator of adverse cardiovascular outcomes in a variety of cardiovascular syndromes [14,16–20].

A chronic total occlusion (CTO) of the coronary artery is a 100% occlusion that is present for over 3 months and is encountered in 15–30% of patients undergoing diagnostic coronary angiography [21]. CTOs are associated with greater ischemia than non-CTO lesions [22]. Collaterals develop over time in response to myocardial ischemia and more profuse development of collaterals has been associated with lower ischemic burden [23]. A previous small study of 35 patients with ≥90% occlusion of the coronary arteries demonstrated an association between higher CD34+ CPC counts and greater collateral circulation [24], but larger studies investigating the role of CPCs in populations with a CTO are lacking. Additionally, whether CPC numbers are directly associated with the ischemic burden, estimated as circulating high sensitivity troponin-I (hsTn-I) levels [25–28], remains unknown. In this prospective cohort of patients with stable CAD with a CTO, we aimed to investigate the association between CD34+ and CD34+/CD133+ CPCs and (a) collateral development and (b) ischemia, estimated with circulating hsTn-I levels. We hypothesized that higher CPC levels will be associated with more robust collateral development and with lower levels of hsTn-I.

2. Methods

2.1. Patients

We studied participants with stable CAD and a CTO enrolled in the Emory Cardiovascular Biobank (NCT00378924), a prospective cohort of patients referred for clinically indicated cardiac catheterization at three Emory healthcare sites in Atlanta, GA between May 2004 and March 2018 [29]. Exclusion criteria included (1) history of cardiac transplantation, (2) history of coronary artery bypass graft surgery, and (3) acute coronary syndrome at presentation.

A CTO was defined as 100% luminal diameter stenosis with absence of antegrade flow on angiography of known or assumed 3 months duration [30]. CTO collateral grade was determined by the Rentrop classification score [31], with a score of 0: having no visible collaterals, 1: having collaterals without filling of the epicardial segment, 2: having partial epicardial artery filling and 3: having complete filling of the epicardial coronary artery. Demographic and clinical data were obtained from questionnaires and medical records included age, sex, race, body mass index (BMI), history of smoking, estimated glomerular filtration rate (eGFR), history of myocardial infarction (MI), history of chronic kidney disease, history of heart failure (HF), left ventricular ejection fraction (LVEF) (defined by most recent transthoracic echocardiography or ventriculography), diabetes mellitus (HbA_1c > 6.5 or treatment with insulin or oral antidiabetic medications), hypertension (systolic blood pressure > 140 mmHg, diastolic blood pressure > 90 mmHg, or treatment with anti-hypertensive medications), dyslipidemia (total cholesterol ≥200 mg/dL, low-density lipoprotein >130 mg/dL, high-density lipoprotein <40 mg/dL or treatment with lipid-lowering medications) as previously described [29]. The study was approved by the Emory University Institutional Review Board. All subjects provided written informed consent.

2.2. CPC Assays

CPCs were measured in peripheral blood samples collected in EDTA tubes before the catheterization procedure. Samples were incubated with fluorochrome-labeled monoclonal antihuman mouse antibodies within 4 hours of collection and enumerated using flow cytometry as CD45^med cells co-expressing CD34 and CD133. 300 μL of peripheral blood was incubated with 7 μL of fluorochrome-labeled isothiocyanate CD34 (BD biosciences, San Jose, California), PerCP-CD45 (BD Biosciences), and 5 μL APC-CD133 (Miltenyi, Bergish Gladbach, Germany) in the dark for 15 minutes [32]. Then, 1.5 mL ammonium chloride lysing buffer was added to lyse red blood cells. 1.5 mL of staining median phosphate buffered saline with 3% heat-inactivated serum and 0.1% sodium azide was added to stop the lysing reaction [32]. Then, 100 μL of Accucheck Counting Beads (Cat#: PCB100; Invitrogen, Carlsbad, California) were added to act as an internal standard for direct estimation of the target cell subset concentrations [32]. At least 2.5 million events were acquired from the cytometer. Flow cytometry data was analyzed using FlowJo software (Tree Star, Ashland, Oregon) with the filter set at CD45^med cells. The exclusion of CD45^bright cells helped to eliminate lymphoblasts and the exclusion of CD45⁻ cells helped eliminate nonhematopoietic progenitors such as mesenchymal or osteoprogenitor cells. CPC counts (CD34+ and CD34+/CD133+) were measured using CD45^med+ filter and are reported as cell counts per mL [32]. In 20 samples that were repeatedly analyzed on 2 occasions by 2 technicians, the percent repeatability coefficients were calculated as the standard deviation of differences between pairs of measurements/mean of measurements x 100. The repeatability coefficients were 2.9% for CD34+ and 4.8% for CD34+/CD133+ [33].

2.3. HsTn-I Assay

Blood samples were collected before cardiac catheterization and were stored at −80 °C. Measurements were performed using the Architect analyzer (Abbott Laboratories North Chicago, IL), with a detection limit of 1.2 ng/L and an interassay coefficient of variation of <10% at 4.7 ng/L.

2.4. Statistical analysis

Patient characteristics were summarized using medians, interquartile ranges (IQR), frequency counts and percentages as appropriate. Baseline differences between patients were assessed using one-way ANOVA and Kruskal-Wallis tests for continuous variables or the chi-squared test for categorical variables. CPC and hsTn-I levels were log₂ transformed prior to analysis. A multivariate ordinal logistic regression was used to evaluate the association between CPC counts and Rentrop collateral grade of 0, 1, 2, or 3. The associations between CPC levels and hsTn-I levels were evaluated using a multivariate linear regression model. Models were adjusted for age [34], sex [35], race (Black vs. non-Black) [36], BMI [37], HF history [19], and chronic kidney disease (CKD) [18] as these have been shown to be correlated to CPC levels. All analyses were performed using SPSS software, Stata/BE 17.0. P-values <0.05 were considered statistically significant.

3. Results

3.1. CPC levels and Rentrop collateral grade

Eighty-nine patients with a CTO and CPC measurements were enrolled (3 with Rentrop collateral grade 0, 22 with grade 1, 37 with grade 2, and 27 with grade 3). Baseline clinical characteristics by collateral development are shown in Table 1. There were no statistically significant associations between the demographic and clinical characteristics with collateral development in patients with CTO in the multivariate ordinal logistic model.

Table 1.

Baseline Characteristics by Rentrop Collateral Classification.

Outcomes	Rentrop Class 0–1 N = 25	Rentrop Class 2 N = 37	Rentrop Class 3 N = 27
Age (IQR)	66.1 (61.4, 77.0)	63.3 (54.3, 71.4)	70.7 (58.1, 82.0)
Sex (Men)	17 (68%)	28 (76%)	19 (70%)
Race (Black)	9 (36%)	12 (32%)	3 (11%)
Hypertension	25 (100%)	35 (95%)	26 (96%)
Dyslipidemia	16 (64%)	29 (78%)	21 (78%)
Diabetes	10 (40%)	19 (51%)	7 (26%)
History of Smoking	18 (72%)	27 (73%)	18 (67%)
BMI (kg/m2) (IQR)	28.4 (23.5, 31.1)	28.1 (23.0, 30.9)	28.2 (25.1, 33.9)
eGFR (mL/min/1.73m2) (IQR)	68.0 (36.1, 80.7)	70.6 (55.3, 97.0)	73.5 (51.9, 93.1)
Chronic Kidney Disease	7 (28%)	8 (22%)	3 (11%)
LVEF (%) (IQR)	55 (45, 60)	55 (35, 60)	55 (55, 60)
History of MI	9 (36%)	12 (32%)	7 (26%)
History of HF	8 (32%)	15 (41%)	6 (22%)
HsTn-I (ng/L) (IQR)	14.4 (8.1, 22.8)	7.0 (3.8, 17.8)	5.1 (3.4, 11.0)
Disease Severity
1 Vessel	8 (32%)	11 (30%)	9 (33%)
2 Vessel	11 (44%)	15 (41%)	10 (37%)
3 Vessel	6 (24%)	11 (30%)	8 (30%)
Circulating Progenitor Cells (cells/mL)
CD34+ (IQR)	1150 (780, 1980)	1170 (710, 1900)	1600 (1010, 2440)
CD34/CD133+ (IQR)	560 (480, 960)	570 (370, 790)	750 (540, 1190)

Values are median (IQR) or n (%).

CTO = chronic total occlusion; BMI = body mass index; IQR = interquartile range; eGFR = estimated glomerular filtration rate; LVEF = left ventricular ejection fraction; MI = myocardial infarction; HF = heart failure; hsTn-I = high sensitivity troponin-I.

There was a nominal trend toward a higher CPC count in those with greater Rentrop collateral grade, Table 1, Fig. 1. After multivariate adjustment with age, sex, race, BMI, HF history, and chronic kidney disease, CPC counts were independent predictors of collateral grade. Thus, in fully adjusted models, every doubling of the CPC count was associated with a ≥ 50% higher Rentrop collateral grade [CD34+ Odds ratio (OR) 1.49 95% CI (0.95, 2.34) P = 0.082] and [CD34+/CD133+ OR 1.57 95% CI (1.05, 2.36) P = 0.028], Table 2.

Fig. 1.

Rentrop Collateral Grade by CPC Level.

Unadjusted bar graph of Rentrop collateral grade and circulating progenitor cell count tertiles. CPC = circulating progenitor cell level.

Table 2.

Relationship between CPC counts and Collateral Development Assessed as the Rentrop Collateral Grade.

	Unadjusted OR (95% CI)	P Value	Model 1 OR (95% CI)	P Value	Model 2 OR (95% CI)	P Value
CD34+ (per 100% increase)	1.51 (0.98, 2.34)	0.064	1.52 (0.98, 2.39)	0.064	1.49 (0.95, 2.34)	0.082
CD34+/CD133+ (per 100% increase)	1.48 (1.01, 2.14)	0.043	1.60 (1.08, 2.39)	0.018	1.57 (1.05, 2.36)	0.028

Relationships were demonstrated using ordinal logistic regression models. CPCs were log₂ transformed. Rentrop collateral grade was scored 0, 1, 2, or 3.

Model 1 adjusted for age, sex, race (black vs. non-black).

Model 2 adjusted for model 1 + BMI, HF history, and chronic kidney disease.

CPC = circulating progenitor cell; OR = odds ratio; CI = confidence interval; BMI = body mass index; HF = heart failure.

3.2. CPC levels and hsTn-I levels

Every doubling of CPC count was also associated with lower hsTn-I levels [for CD34+ cells, β = −0.36 95% CI (−0.51, −0.16) P = 0.002] and [for CD34+/CD133+, β = −0.26 95% CI (−0.42, −0.07) P = 0.011]. This remained significant after full adjustment [CD34+ β −0.35 95% CI (−0.49, −0.15) P = 0.002] and [CD34+/CD133+ β −0.27 95% CI (−0.43, −0.08) P = 0.009], Table 3.

Table 3.

Relationship between CPC counts and HsTn-I Levels.

	Unadjusted β (95% CI)	P Value	Adjusted Model 1 β (95% CI)	P Value	Adjusted Model 2 β (95% CI)	P Value
CD34+ (per 100% increase)	−0.36 (−0.51, −0.16)	0.002	−0.36 (−0.51, −0.15)	0.002	−0.35 (−0.49, −0.15)	0.002
CD34+/CD133+ (per 100% increase)	−0.26 (−0.42, −0.07)	0.011	−0.28 (−0.43, −0.08)	0.008	−0.27 (−0.43, −0.08)	0.009

Relationships were demonstrated using linear regression models. CPCs and hsTn-I were log₂ transformed.

Model 1 adjusted for age, sex, race (black vs. non-black).

Model 2 adjusted for model 1 + BMI, HF history, and chronic kidney disease.

CPC = circulating progenitor cell; HsTn-I = high sensitivity troponin-I; CI = confidence interval; BMI = body mass index; HF = heart failure.

In the unadjusted model, a higher Rentrop collateral grade was also associated with a lower hsTn-I level. Each unit increase in Rentrop collateral grade was associated with a 30% lower hsTn-I level [β = −0.30; 95% CI (−0.50, −0.02) P = 0.040].

4. Discussion

Novel findings of our study are that a higher CPC count was independently associated with greater collateralization, indicating the important role of CPCs in collateral formation in the presence of a CTO. No other demographic or clinical characteristic was a predictor of collateralization. A higher CPC count was also associated with a lower ischemic burden, determined as lower levels of hsTn-I.

4.1. CPC levels and collaterals

Ischemia from myocardial injury triggers the upregulation of hypoxia-inducible factor 1-alpha that stimulates expression of stromal cell derived factor-1 to permit homing of CPCs to site of injury [38–40]. Hypoxia-inducible factor 1-alpha also stimulates nitric oxide-dependent matrix metallopeptidase 9 production that promotes synthesis of vascular endothelial growth factor (VEGF) that in turn stimulates CPC-dependent angiogenesis [38,41]. Although differentiation of CPCs into the newly formed vasculature has been suggested as one mechanism linking CPCs to angiogenesis [12,13,42,43], the paracrine actions of CPCs by the transfer of exosomes and activation of local angiogenic pathways are likely the predominant mechanisms [44–47].

The association between CPC levels and collateral development has not been well established previously. Although one small study of 35 patients showed an association of high CD34+ cell counts with robust collateral circulation in obstructive CAD ≥ 90% stenosis [24], there have not been larger studies, to our knowledge, to support his relationship. Additionally, this relationship has not been previously described in patients with CTO. In patients with CTO, another study showed a positive correlation between the proportion of the homing signal C-X-C motif chemokine receptor 4 (CXCR4)-expressing CD34+ cells and the Rentrop collateral score [48]. Our data supplements these findings by showing a positive correlation between the absolute number of CD34+ and CD34/CD133+ cell levels and the extent of collateral circulation, suggesting that endogenous regenerative capacity has an important role in coronary collateral development in patients with CTO. Interestingly, no other characteristics contributed to collateral development.

Why some individuals with CAD have fewer CPCs compared to others is not well understood. CPCs are lower in women compared to men [35], in Black compared to white participants [36] and with persistent injury from exposure to cardiovascular risk factors [16], aging [34], renal insufficiency [18], advanced peripheral artery disease [17] and heart failure [19]. Lower CPC counts are a reflection of lower progenitor cell levels in the bone marrow and, additionally, may imply abnormalities in the mobilization of the progenitor cells from the bone marrow [14,15]. In settings of ischemia, some studies have shown a transient lowering of CPC counts [49], implying homing of CPCs into the injured myocardium [50]. Although not previously studied in settings of chronic ischemia, our data demonstrates that those with endogenously high CPC counts had more robust collaterals than those with lower counts.

4.2. CPCs, collaterals and hsTn-I

Another novel finding of this study is the association between higher CPC levels and lower ischemic burden, measured as hsTn-I levels. CTO lesions are associated with high levels of ischemia [22,51–54]. Patients with CTO and poorly developed collaterals have higher levels of ischemia than those with well-developed collaterals as demonstrated with non-invasive imaging [23] and circulating hsTn-I levels [55]. Our data supports these observations, as the extent of collateralization was inversely associated with hsTn-I levels. Furthermore, we found that a higher CPC count was associated with lower hsTn-I levels, a novel finding, suggesting that endogenous regenerative capacity, through enhanced collateral development reduces the ischemic burden of a CTO. Additionally, two phase 2, single-arm clinical trials have shown that among patients with nonobstructive CAD, patients that received CD34+ cell therapy showed improvement in microvascular coronary blood flow and decreased angina after CD34+ cell therapy [56,57]. Thus, decreased ischemic burden, may also be mediated by angiogenesis at the microvascular level.

It has been well established that higher CPC counts are associated with lower adverse cardiovascular event rates in patients with CAD [16,58]. Similarly, high hsTn-I levels are associated with higher risk of incident adverse events [59]. Implications from our findings are that at least partly, individuals with high CPC counts have lower adverse event rates because of lower levels of chronic ischemia possibly secondary to greater collateral development [60,61].

4.3. Limitations

Strengths of our study include enrollment of a diverse population including women and Black participants. We used high-throughput technology (flow cytometry) for quantification of CPCs by standardized and reproducible assay techniques. Our study had a few limitations. Given the limited sample size of 89 patients, our findings may not be fully representative of the entire population. We only studied patients with clinically indicated cardiac catheterization, and therefore our results may not apply to the general population. We also did not measure the functional capacity of CPCs, but previous studies have shown correlations between CPC numbers and their proliferative potential [62]. The cross-sectional nature of this analysis does not prove causation between CPC counts and collateralization, and thus longitudinal data is required. Additionally, we evaluated the degree of collateralization by Rentrop scoring system instead of functional tests such as the collateral flow index.

5. Conclusion

In the presence of a CTO, individuals with higher CPC counts have greater collateral development and lower ischemic burden.

Abbreviations:

CPC

Circulating progenitor cell

CAD

Coronary artery disease

CTO

Chronic total occlusion

CD

Cluster of differentiation

BMI

Body mass index

eGFR

Estimated glomerular filtration rate

MI

Myocardial infarction

HF

Heart failure

LVEF

Left ventricular ejection fraction

CI

Confidence interval