| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;83:598-605
© 2007 The Society of Thoracic Surgeons
Departments of Cardiothoracic Surgery, and Cardiac and Vascular Sciences, St. George’s Hospital Medical School, London, United Kingdom
Accepted for publication September 25, 2006.
* Address correspondence to Dr Jahangiri, Department of Cardiothoracic Surgery, St. George’s Hospital, Blackshaw Road, London, SW17 0QT, UK (Email: marjan.jahangiri{at}stgeorges.nhs.uk).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: One hundred and ten patients were studied: 54 elective CABG (30 on-pump, 24 off-pump); 23 urgent CABG; and 33 non-CABG. The EPC functional status was assessed using the colony forming unit assay (EPC-CFU) and plasma levels of granulocyte colony-stimulating factor (G-CSF), stromal cell-derived factor 1-
, matrix metalloproteinase, and vascular endothelial growth factor were assessed by enzyme-linked immunosorbent assay. Samples were taken preoperatively and on days 1 and 5 after surgery.
RESULTS: Patients requiring urgent CABG and non-CABG patients had significantly higher numbers of EPC-CFU prior to surgery than elective CABG patients. All elective patients showed a significant increase in postoperative EPC-CFU levels: on-pump CABG 10.4 ± 3.8 to 53.9 ± 11.9, p = 0.001; off-pump CABG 9.5 ± 3.5 to 65.7 ± 17.3, p = 0.006; non-CABG 23.5 ± 6.8 to 84.6 ± 27.2, p = 0.05. The postoperative EPC rise in elective patients correlated with plasma G-CSF levels (r = 0.387, p < 0.01). Urgent patients demonstrated a significant increase in G-CSF levels but this was not associated with an increase in EPC-CFU level.
CONCLUSIONS: Patients undergoing elective cardiac surgery demonstrated an increase in EPC-CFU postoperatively, which correlated with increased plasma G-CSF level. Urgent patients did not have an increase in EPC-CFU despite a plasma G-CSF rise. Endogenously mobilized EPCs present a potential therapeutic target.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Levels of circulating EPCs are known to be reduced in patients with cardiovascular disease [6], or patients with risk factors for cardiovascular disorders [7]. These levels have been shown to be predictive of cardiovascular mortality in patients with coronary artery disease [8, 9].
Endothelial progenitor cells have been considered a target for the development of cellular therapies; however, those with most need of such therapy have the lowest number of EPCs in their circulating blood. Hence, current options are ex vivo culture expansion of EPCs prior to reinfusion or stimulation of EPC release from the bone marrow by cytokines such as granulocyte colony stimulating factor (G-CSF) [10, 11].
Endothelial progenitor cell therapy has been used with some success in patients with peripheral vascular disease [12, 13] and in patients after acute myocardial infarction [14]. However, the optimal therapeutic strategy for cellular therapy in patients with chronic ischemic heart disease is unknown.
Evidence exists that EPC number may increase after coronary artery bypass graft (CABG) surgery with the use of cardiopulmonary bypass (CPB) [15, 16], but there is little known about EPC response to off-pump surgery. We have found no evidence in the literature describing the role of EPCs after valve surgery or urgent CABG.
Prior to development of cellular therapeutic strategies for surgical use, knowledge of endogenous cytokine release during surgery and the effect on progenitor cell mobilization is essential. Our choice of cytokines was based on previously published work examining the mobilization of EPCs especially in studies on cardiac surgical patients. Granulocyte colony stimulating factor is a well-known mobilizing cytokine used extensively in clinical settings to mobilize hematopoietic stem cells, and it has also been used clinically in patients with ischemic heart disease [10, 11]. The vascular endothelial growth factor (VEGF) was linked to EPC release in the first description of EPC release after cardiac surgery in burn patients by Gill [16], and also examined by Scheubel and colleagues [15] after on-pump CABG. Stromal derived factor-1 (SDF-1) has been implicated in progenitor cell trafficking [17] and recently described in relation to EPC release after on-pump CABG.
The full mechanism by which mobilization of EPCs occurs is not fully understood. It is thought that cytokines such as G-CSF, SDF-1, or VEGF may act to release EPCs from the bone marrow in a matrix metalloproteinase-9 (MMP-9) dependent mechanism [3]. Hence, we chose to investigate these cytokines in our study. The aim of this study was to characterize cytokine and EPC response not only to elective CABG surgery with CPB but also off-pump CABG, valve surgery, and urgent CABG.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Urgent patients were defined as patients referred for CABG as inpatients that had suffered a troponin positive event or had unstable angina within 30 days prior to surgery. They received surgery on routine operating lists and were all performed during normal working hours.
Elective patients discontinued antiplatelet agents and angiotensin-converting enzyme inhibitors at least five days prior to surgery. Statin therapy was continued until the day before surgery and reinstituted in the evening of the first postoperative day. Venous blood samples were collected preoperatively and on days 1 and 5 after surgery. Preoperative and day 1 samples were collected through a central venous catheter with aspiration of at least 10 mL dead space prior to sample collection. Day 5 samples were taken by direct venous puncture.
Blood was collected in ethylenediaminetetraacetic acid (EDTA) tubes and processed within three hours of collection. The EPC-colony forming unit (CFU) assay and the EPC culture assay were performed within three hours of sample collection. Plasma for subsequent enzyme-linked immunosorbent assay (ELISA) experiments was aliquoted and frozen for future use within three hours of sample collection. Blood for plasma extraction was collected in EDTA tubes and then centrifuged at 3,000 g for 10 minutes; after a period of a further 10 minutes, aliquots of 500 microliters were then obtained and stored at –80°C until required.
Cardiovascular Risk Factors
A cumulative number of cardiovascular risk factors were identified for each elective patient. These included the following: presence of diabetes mellitus (treated by insulin or oral hypoglycemic agents), family history of ischemic heart disease in first degree relative under 65 years, history of smoking within 20 years of procedure, hypertension, and hypercholesterolemia.
Operative Technique
All patients received a general anesthetic as per our institutional protocol. On-pump patients were managed according to a standard protocol, namely mild systemic hypothermia (32°C) with perfusion pressure maintained above 60 mm Hg. Cardiac arrest was achieved with antegrade cold blood-based cardioplegic solution. Off-pump surgery was performed using the CTS stabilizer (Cardio Thoracic Systems Inc, Cupertino, CA).
EPC-CFU Assay
Venous blood samples were collected as described above and the assay performed as described previously [7, 19–21]. Peripheral blood mononuclear cells were isolated by density gradient centrifugation over Lymphoprep (Axis-Shield, Oslo, Norway).
After three wash cycles the mononuclear cells were resuspended in EPC culture medium (M199 with 20% fetal calf serum [FCS] and antibiotics) and then plated on fibronectin-coated 6-well plates at a concentration of five million cells per well. This step was designed to remove mature circulating endothelial cells, which rapidly adhere to fibronectin. After culture for 48 hours, nonadherent cells were then aspirated and counted. Cells nonadherent and hence aspirated along with the supernatant were then plated on fibronectin-coated 24-well plates at a concentration of one million cells per well. On day 5 of the assay, the medium was changed and on day 7 the endothelial colonies were counted manually.
Strict guidelines were followed to ensure consistent counting of EPC colonies. Colonies were only counted as EPC-CFUs if they consisted of more than 50 cells and contained a core of rounded cells with flat, spindle-shaped cells emanating from the periphery (Fig 1). Colonies were counted in a minimum of 3 wells and an average was then recorded. The investigator and a member of senior laboratory staff who was blinded to the patients’ clinical status counted colonies. Reproducibility was assessed over 10 samples comparing colony counts by the two individuals. Coefficient of variance was less than 10% in each case. Selected colonies were stained with DiI low-density lipoprotein (LDL), lectin, kinase domain region (KDR), and AC133 to confirm phenotype.
|
The nonadherent cells were then removed by thorough washing with phosphate-buffered saline (PBS). Fresh medium was then added and cells cultured for another 48 hours at 37°C and 5% CO2.
On the seventh day of culture, medium was aspirated and the adherent cells were again thoroughly washed with PBS. Serum-free RPMI 1640, containing 2.0 µg/mL DiI-Ac-LDL, was then added and the cells incubated at 37°C for one hour. The medium was then aspirated and the cells washed twice with PBS to remove free Dil-Ac-LDL and then fixed with a 4% formalin-PBS solution. Cells were then incubated with 10 ug/mL fluorescein isothiocyanate (FITC) labeled Ulex europaeus agglutinin (UEA; Lectin, Sigma) at 37°C for one hour. After a final wash with PBS the cell nuclei were then counter-stained with 4',6-diamidino-2-phenylindole (DAPI).
Chamber slides were then examined under a fluorescence microscope with the filter set for rhodamine, FITC, and DAPI. Double positive stained cells were counted as EPCs and the total number counted per slide. This counting was undertaken by a single trained member of the senior laboratory staff blinded to the clinical details of the patient. This assay was performed on a subgroup of the study population (n = 49) once the technique had been established.
ELISA Experiments
All ELISA protocols were carried out at room temperature on freshly thawed plasma samples. Plasma G-CSF, SDF-1, VEGF, and MMP-9 levels were determined using commercially available ELISA kits (Quantikine, R&D Systems, Abingdon, UK). The concentration of all cytokines was determined by comparison with a standard curve, as per manufacturer’s instruction.
Statistical Analysis
Values are expressed as mean ± standard error of the mean or number (%). Statistical analysis was performed using SPSS version 13 (SPSS Inc, Chicago, IL). Comparisons of EPC numbers and cytokines within and between groups were made using the Mann-Whitney U test to avoid assumptions of normal distribution of data. Bivariate correlations were assessed using the Spearman correlation coefficient. A p value less than 0.05 was considered significant.
Predictors of day 1 EPC-CFU were identified by univariate linear regression and then significant factors were entered into a multiple linear regression model. Only day 1 EPC and cytokine values were used in this analysis.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
Mobilization of EPCs
Figure 2
illustrates the EPC-CFU levels in the four patient groups over the three time points of the study. It is clear that the three elective patient groups all demonstrated a significant increase in EPC-CFU after surgery. However, the urgent patients who had all recently suffered troponin positive events had a higher preoperative EPC-CFU score but did not demonstrate an increase after surgery. This clearly demonstrates an endogenous mobilization of functionally active EPCs into the circulation by patients who had documented coronary artery disease (CABG patients) and associated lower preoperative EPC levels than patients with documented absence of coronary artery disease (non-CABG patients).
|
Plasma Cytokines
Plasma cytokine profiles over the three time points are shown in Figure 3. It is clear that plasma G-CSF increased significantly in all groups after surgery when compared with preoperative levels. There were no significant differences in day 1 G-CSF levels among the groups, despite a trend toward higher levels in the non-CABG group. In the elective patients, day 1 G-CSF levels correlated with day 1 colony levels (r = 0.387, p < 0.01).
|
The MMP-9 levels increased in all elective groups on day 1 compared with preoperative levels, but not in the urgent group. The MMP-9 levels did not correlate with EPC-CFU levels at any time point. The SDF-1 levels decreased significantly on day 1 postoperatively but returned to preoperative levels by day 5.
Table 3 shows the predictors of day 1 EPC-CFU as defined by univariate and multiple linear regression. Day 1 G-CSF was the only factor identified on univariate analysis, which maintained its significance in the multiple linear regression model. This supports the hypothesis that G-CSF is a factor in the mobilization of EPCs after cardiac surgery.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Our preoperative results agree with previous studies suggesting patients with coronary artery disease have reduced EPC function and that this correlates with the number of risk factors for cardiovascular disease [6, 7]. Specifically, EPC-CFU numbers in patients with low risk for cardiovascular disease or healthy volunteers were measured at a mean of 20 to 25 colonies per well in a number of studies using the CFU assay [7, 10, 20, 21]. In these studies, known vascular disease or dysfunction resulted in a reduced EPC-CFU number of 5 to 10 colonies per well. We have also shown that patients with unstable angina have higher numbers of EPC colonies than those with chronic stable angina [19]. These baseline results confirm consistency of the EPC-CFU assay in our laboratory. It also demonstrates that our elective CABG patients did have a reduced level of circulating EPCs and that the patients undergoing non-CABG surgery had normal levels of circulating EPCs when compared with healthy volunteers from previous studies.
We have shown that EPC colony formation, a measure of functional activity of EPCs rather than a measure of number, increases after CABG whether performed on- or off-pump. We have also demonstrated that EPC colony formation is increased after non-CABG cardiac surgery.
Other authors have reported an increase in the number of EPCs after CABG with CPB. Gill and colleagues [16] documented a rise in KDR positive cells after CABG in seven patients and Scheubel and colleagues [15] described an increase in CD34/AC133 positive cells in 50 patients undergoing on-pump CABG. Ruel and colleagues [23] recently described a rise in the number of EPCs measured by cell culture in 10 patients undergoing off-pump and 10 patients undergoing on-pump CABG, although they described reduced function in the on-pump patients. We did not demonstrate a reduction of EPC colony formation in relation to CPB. As the EPC-CFU assay assesses EPC viability and proliferative ability, this suggests no loss of function in relation to CPB. As controversy still exists over the exact definition of circulating EPCs, comparisons between studies using different methods must be guarded.
In a recent study comparing the methods of EPC assessment it was concluded that EPC-CFU is a good measure of EPC function [24], and as EPC-CFU has been correlated to a clinical assessment of endothelial function [7] and to cardiovascular mortality [8], we are confident that these cells represent a cardiovascularly active subpopulation of circulating EPCs.
Cytokines and EPC Mobilization
This study documents the G-CSF response to cardiac surgery and its correlation with EPC-CFU, and suggests that G-CSF plays a role in the mobilization of EPCs in the acute setting. This finding is in agreement with other published studies [10, 11], which have suggested that G-CSF can mobilize EPCs in patients with ischemic heart disease by subcutaneous injection. The G-CSF has also been shown to be endogenously released in patients with acute myocardial infarction, and significantly correlated to mobilized CD34+ cells [25]. We have shown no correlation between SDF-1 and EPC levels postoperatively, which is in agreement with the studies after myocardial infarction [25].
Recently, Mieno and colleagues [18] published data describing the role of SDF-1 in the mobilization of EPCs after CABG, suggesting an increase in SDF-1 four hours after surgery which had returned to preoperative levels by day 4. We found a reduction in SDF-1 on day 1 across all four of our patient groups, suggesting if SDF-1 is systemically involved in EPC mobilization it is an early response in the first few hours after surgery. Mieno and colleagues reported increased G-CSF postoperatively in similar levels to our data and similar VEGF levels to our study, with no rise early after surgery but significant rises on day 4.
We hypothesized that acute ischemia results in G-CSF release, which is involved in EPC mobilization. The exact mechanism for such release is unknown; it is possible that other cytokines, such as SDF-1, are implicated prior to G-CSF release. Further time course analysis of mobilizing cytokines is warranted to elucidate the exact mechanism of mobilization.
In the urgent patient group, we hypothesized that EPCs were released after the acute ischemic event prior to surgery. In most patients there were at least five days between the troponin positive event and surgery. This time scale would explain why G-CSF levels were not yet raised on the day of surgery when preoperative blood samples were taken. These data do not support that G-CSF is the sole factor responsible for mobilizing EPCs after surgery; however, it provides further evidence that it has a significant role.
EPCs and Acute Cardiovascular Events
It is well-documented that EPCs are mobilized in response to acute cardiac events such as unstable angina [19] and acute myocardial infarction [26–28], with G-CSF implicated in this response. We have demonstrated that after an acute coronary event, EPCs are indeed higher than in patients with chronic stable angina. We have also examined the effect of a further G-CSF stimulus in such patients in our subgroup of patients undergoing urgent CABG. We have shown that in this situation a G-CSF stimulus in the form of surgery does not significantly increase EPC level from preoperative values. This suggests that G-CSF influences a pool of EPCs available for quick release, but that this pool is of finite number and takes time to replenish.
Clinical Relevance
G-CSF mobilizes EPCs into the circulation but this alone is insufficient to offer clinical benefit to patients with ischemic heart disease [10, 11, 29], potentially due to the deficient homing mechanisms described in the EPCs of patients with cardiovascular disease [30, 31]. As G-CSF is endogenously released during cardiac surgery, with functionally active EPCs mobilized into the circulation, this offers a potential therapeutic target. However, until we understand more about the homing characteristics of these mobilized cells, whether they will offer competition to the strategy of culture expansion and then reinfusion of EPCs remains unknown.
We hypothesized that data from this study allow us to explore the possibility of targeting these mobilized EPCs by using a homing cytokine such as SDF-1, which would be injected into areas of damaged myocardium at the time of surgery. This would allow EPC therapy to be targeted to dysfunctional areas without the need for a preoperative culture expansion of a patient’s cells. Indeed, such a strategy has already been attempted in an animal model with encouraging results, with SDF-1 used as a homing cytokine and G-CSF used to mobilize EPCs [32]. Our data do not support clinical use of such a strategy yet, but they suggest that such a strategy may be possible.
In summary, this study demonstrates that functionally active EPCs are increased in the circulation after all types of cardiac surgery and correlate with G-CSF levels. However, before they can be harnessed for therapeutic benefit better understanding of the homing mechanisms of the mobilized cells is needed.
Study Limitations
We designed the study specifically to describe the EPC and cytokine response to a variety of cardiac surgical interventions. This has been achieved but incorporates elements of variability between patients, which is difficult to control for with each patient’s preoperative sample having to act as his or her own control.
There is a difference in terms of preoperative medication among the groups of patients, most prominently the low level of statin therapy in the non-CABG group and the lack of antiplatelet therapy in all the elective patients. While these differences in medical therapy are potentially confounding factors, we suggest that the role of such therapies is minimal when compared with the surgical procedures undertaken.
There is controversy over the exact definition of an EPC and hence controversy on the gold standard for assessment of number and function. There are other laboratory protocols that could have been used, including migration assays and in vitro angiogenesis assays. We chose the CFU assay due to its correlation to endothelial function and cardiovascular prognosis.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Herrmann, A. M. Abarbanell, B. R. Weil, Y. Wang, M. Wang, J. Tan, and D. R. Meldrum Cell-based therapy for ischemic heart disease: a clinical update. Ann. Thorac. Surg., November 1, 2009; 88(5): 1714 - 1722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Dotsenko and M. Jahangiri Endogenous stem cells in patients undergoing coronary artery bypass graft surgery Eur. J. Cardiothorac. Surg., September 1, 2009; 36(3): 563 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Xu Stem Cells and Transplant Arteriosclerosis Circ. Res., May 9, 2008; 102(9): 1011 - 1024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. E. Thomas, R. Redgrave, M. S. Cunnington, P. Avery, B. D. Keavney, and H. M. Arthur Circulating Endothelial Progenitor Cells Exhibit Diurnal Variation Arterioscler Thromb Vasc Biol, March 1, 2008; 28(3): e21 - e22. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
), p = 0.03 and p = 0.001, respectively. (Light gray columns = preoperative; dark gray columns = day 1; black columns = day 5.)