INTRODUCTION

The primary objective of coronary artery bypass surgery (CABS) is to accomplish full revascularization with conduits that will remain patent for the lifetime of the patient. Atherosclerotic factors, such as diabetes mellitus (DM) and older age, can influence graft patency. In addition to these, activities of platelet molecules may play an important role on the initiation and progression of thrombus formation and the inflammatory processes of atherosclerosis^[1-3]. Platelet distribution width (PDW) directly measures the variability in platelet size. And larger platelets are both metabolically and enzymatically more active than smaller ones. When there is variation in platelet size (anisocytosis), increased PDW may be evident. Also, platelets with a large size may indicate the presence of platelet aggregates, which may influence the occlusion of a saphenous vein graft (SVG)^[4]

The objective of this study was to identify predictive factors for SVG failure in patients undergoing coronary artery surgery (CAS). To do this, all clinical hematological variables were evaluated in the same statistical pool.

METHODS

A total of 1950 consecutive patients underwent CABS in our institute from November 2010 to February 2013. Our study consisted of 284 patients who had undergone coronary computed tomography angiography (CCTA) due to suspicion of progressive coronary artery disease (CAD). Although all patients in our study were under optimal medical treatment, every patient with recurrence of angina symptoms or findings of ischemic changes from noninvasive imaging modalities, such as cardiovascular exercise stress tests or myocardial perfusion scintigraphy, underwent coronary angiography after their surgery.

Of the total 784 distal anastomoses in the patients of our group study, 131 SVG (in 109 patients) were occluded. Sequential and T grafting were used in 13 of 109 patients with at least one graft occlusion. Repeated revascularization was performed in only 13 patients, because their lesions were suitable for treatment with percutaneous coronary intervention.

We reviewed the preoperative and postoperative coronary angiograms of these patients retrospectively. Groups were created according to the presence of at least one SVG failure. In our hospital, patients with stable angina symptoms and/or positive stress test results routinely undergo CCTA examination first. Therefore, conventional coronary angiograms of patients with occluded grafts based on CCTA are elective.

Patients whose SVG were harvested endoscopically, who had previous atrial arrhythmia, infection, inflammatory disease, acute coronary syndrome, severe hepatic diseases, malignant neoplasms, renal failure (serum creatinine level > 2.0 mg/dL), recent history of blood transfusion, hematological disorders such as severe anemia (hemoglobin level < 10 g/dL), concomitant valvular surgery, emergency, and congenital heart disease, without regular follow-up period, and without taking medical treatment consisted of statin or use of beta-blockers after CABS were excluded from the study. Additionally, patients who underwent off-pump surgery or redo surgery were excluded.

Written informed consents were obtained from all patients before enrollment. The study was also approved by the local ethics committee and the institutional review board of the Istanbul Mehmet Akif Ersoy Thoracic and Cardiovascular Surgery Training and Research Hospital (code: 2018-12). The study was consistent with the Declaration of Helsinki.

RESULTS

Baseline Characteristics

Two hundred and eighty-four patients (223 men, 61 women) underwent CAS. The mean age was 62 years (range: 31 to 85 years). A total of 109 patients (90 men, 19 women) with mean age of 63.5 years were included in the SVG occluded group and 175 patients (133 men, 42 women) with mean age of 57.6 years were included in the SVG patent group. The rate of SVG failure was not affected by sex (P=0.190). Hematologic parameters and the clinical characteristics of the patients and a comparison of cardiovascular risk factors are summarized in Tables 1 and 2. In the univariate analysis, older age, smoking, EF value, medical history of myocardial infarction (MI), DM, postoperative AF, cerebrovascular events, and preoperative SYNTAX score were statistically significantly different between the two groups. The EF value was lower in the SVG occluded group than in the SVG patent group (P<0.000). When the two groups were compared with respect to laboratory results, serum creatinine level, mean platelet volume, and PDW were statistically significantly higher in patients with occlusion than those without it.

Table 1 - Patients’ basic clinical and laboratory features.

		SVG occluded group		SVG patent group		P value
		Mean ± SD/n-%		Mean ± SD/n-%
Age (years)		63.5±7.9		57.6±8.5		0.000 m
Sex	Female, n (%)	19	17.4%	42	24%	0.190 X2
	Male, n (%)	90	82.6%	133	76%
BMI (kg/m2)		27.9±3.7		28.8±4.3		0.158 m
DM, n (%)		75	68.8%	69	39.4%	0.000 X2
Hypertension, n (%)		64	58.7%	88	50.3%	0.166 X2
Smoking		71	65.1%	58	33.1%	0.000 X2
Peripheral arterial disease, n (%)		8	7.3%	12	6.9%	0.877 X2
COPD, n (%)		22	20.2%	28	16%	0.368 X2
Previous MI, n (%)		64	58.7%	65	37.1%	0.000 X2
Previous PCI, n (%)		42	38.5%	64	36.6%	0.740 X2
Ejection fraction (%)		45±8.9		56.4±7		0.000 m
Preoperative SYNTAX score		27.6±9.2		23.5±10.1		0.000 m

m Mann-Whitney U Test;

X2 Chi-square test.

BMI=body mass index;

COPD=chronic obstructive pulmonary disease;

DM=diabetes mellitus;

MI=myocardial infarction;

PCI=percutaneous coronary intervention;

SD=standard deviation;

SVG=saphenous vein graft

Table 1 - Patients’ basic clinical and laboratory features.

Table 2 - Blood test results before the CT coronary angiogram.

	SVG occluded group	SVG patent group	P value
	Mean ± SD	Mean ± SD
Total cholesterol level (mg/dL)	209.2±68.1	200±55.2	0.404 m
HDL (mg/dL)	38.5±10.5	39.4±12.5	0.753 m
LDL (mg/dL)	134.7±43.6	128.5±45.2	0.254 m
FPG (mmol/L)	139.9±62.2	136.6±57.8	0.867 m
HbA1C	7.2±2.2	6.8±1.7	0.352 m
Hb (gr/dL)	13.8±1.7	13.5±1.6	0.232 m
MPV (fL)	8.9±1	8.5±1.1	0.002 m
WBC (x 109/L)	8.1±1.9	7.8±1.9	0.290 m
Lymphocyte (x 109/L)	2.3±0.9	2.4±0.8	0.328 m
Neutrophil (x 109/L)	4.9±1.6	4.8±1.5	0.712 m
MCV (fL)	86.2±5.9	86.2±6.5	0.805 m
Plt (x1000/mL)	240.9±78	259±87.8	0.054 m
PDW	16.9±2	15.3±1.2	0.000 m
RDW	13.9±1.6	13.5±1.3	0.056 m
Uric acid (mg/dL)	5.8±1.8	5.3±1.4	0.269 m

m Mann-Whitney U Test.

CT=computed tomography;

FPG=fasting plasma glucose;

Hb=hemoglobin;

HDL=high-density lipoprotein;

LDL=low-density lipoprotein;

MCV=mean corpuscular volume;

MPV=mean platelet volume;

PDW=platelet distribution width;

Plt=platelet;

RDW=reticulocyte distribution width;

SD=standard deviation;

SVG=saphenous vein graft;

WBC=white blood cells

Table 2 - Blood test results before the CT coronary angiogram.

Operative and Postoperative Data

The numbers of bypass grafts and use of red blood cell units in the SVG occluded group were significantly higher than in the SVG patent group (Table 3).

Table 3 - Operative and postoperative data.

	SVG occluded group		SVG patent group		P value
	Mean ± SD		Mean ± SD
Number of bypass grafts (n)	3.5±0.8		3.1±0.7		0.000 m
Cardiopulmonary bypass time (min)	108.2±51.3		91.2±32.3		0.011 m
Cross-clamp time (min)	61.6±44.9		49.2±23.5		0.051 m
Blood transfusion amount (units)	4±3		2.9±2.2		0.001 m
Follow-up time (months)	49.96±3.93		48.89±4.06		0.072 m
Interval from operation to CT angiogram (months)	40.0±3.9		39±3.6		0.242 m
LAD-LITA	108	99.1%	170	97.1%	0.269 X2
Occluded LITA	4	3.7%	3	1.7%	0.301 X2
LMCA	17	15.6%	21	12%	0.387 X2
Clopidogrel	15	13.8%	12	6.9%	0.054 X2
Sequential anastomosis	9	8.3%	8	4.6%	0.203 X2
T graft	4	3.7%	1	0.6%	0.073 X2
POAF	31	28.4%	32	18.3%	0.045 X2
Cerebrovascular incident, n (%)	7	6.4%	2	1.1%	0.013 X2
Length of stay in hospital (days)	9.6±6.3		8.5±2.9		0.223 m

m Mann-Whitney U Test;

X2 Chi-square test.

CT=computed tomography;

LAD-LITA=left anterior descending-left internal thoracic artery;

LITA=left internal thoracic artery;

LMCA=left main coronary artery;

POAF=postoperative atrial fibrillation;

SD=standard deviation;

SVG=saphenous vein graft

Table 3 - Operative and postoperative data.

Cumulative Analysis

In a multivariate logistic regression model, a significant association was found between graft occlusion and older age, DM, PDW, history of previous MI, and EF (Table 4).

Table 4 - Logistic regression model of risk factors for saphenous vein graft failure.

	Univariable model					Multivariable model
	OR	95% confidence interval			P	OR	95% confidence interval			P
Age	1.10	1.06	-	1.13	0.000
Blood transfusion amount	1.19	1.08	-	1.31	0.000
Smoking	3.77	2.28	-	6.24	0.000	3.17	1.14	-	8.81	0.027
Diabetes mellitus	3.39	2.04	-	5.62	0.000	3.46	1.21	-	9.83	0.020
Prevıous MI	2.41	1.48	-	3.93	0.000
POAF	1.78	1.01	-	3.13	0.047
Cerebrovascular incident	6.00	1.22	-	29.41	0.027
MPV	1.42	1.13	-	1.80	0.003
PDW	2.38	1.73	-	3.27	0.000	3.53	1.97	-	6.35	0.000
Ejection fraction	0.85	0.82	-	0.88	0.000	0.84	0.79	-	0.89	0.000
Creatinine	72.2	20.5	-	255.1	0.000
Cardiopulmonary bypass time	1.01	1	-	1.02	0.009
Preoperative SYNTAX Score	1.04	1.02	-	1.07	0.001	1.07	1.01	-	1.13	0.013
Number of grafts	1.82	1.32	-	2.50	0.000

MI=myocardial infarction;

MPV=mean platelet volume;

OR=odds ratio;

PDW=platelet distribution width;

POA=postoperative atrial fibrillation

Table 4 - Logistic regression model of risk factors for saphenous vein graft failure.

Receiver Operator Characteristic Curve Analysis

The area under the ROC curve (95% confidence interval) for PDW as a predictor of SVG failure was 0.765 (0.70-0.82, P<0.000) (Figures 1 and 2). The PDW cut-off point was > 16 for a sensitivity of 57.5% and specificity of 84%.

Fig. 1 - Receiver operator characteristic curve of platelet distribution width (PDW) in detecting patients with saphenous vein graft failure.

Download

Fig. 1 - Receiver operator characteristic curve of platelet distribution width (PDW) in detecting patients with saphenous vein graft failure.

Fig. 2 - Box-plot diagram of platelet distribution width (PDW) by group.

Download

Fig. 2 - Box-plot diagram of platelet distribution width (PDW) by group.

DISCUSSION

In this study, we investigated the relationship between SVG failure and the clinical hematological parameters of patients who had undergone CABS. We found that age, DM, history of previous MI, and PDW were significantly associated with vein graft failure. A history of previous MI and protected left ventricular functions yields similar results. In other words, MI can cause decreased ventricular function. Apart from these factors, our study demonstrated that a good EF was a protective factor against SVG failure. Older age and DM can be thought of as preoperative and non-specific factors. Therefore, it is difficult to estimate post-operative SVG failure. However, PDW is both a specific and pre-angiographic predictive parameter for SVG failure after the procedure and PDW is a more specific marker of platelet activation than other platelet indices, such as mean platelet volume, because PDW does not increase during simple platelet swelling^[6]. For these reasons, PDW may provide valuable predictive information regarding the potential for SVG failure following CABS.

The occlusion or stenosis of SVG within the first year after the operation is attributed to surgical technical errors, thrombosis, and occlusive intimal hyperplasia^[7]. Beyond one year, inflammation that causes increased platelet activation plays a major role in the development of atherosclerosis^[2]. Platelet count, mean platelet volume, and PDW may reflect the degree of platelet activation. The relationship between atherosclerosis and increments in platelet activity is well known.

However, there is not a great deal of data regarding the relationship between PDW and SVG failure. Moreover, contrasting results have been reported^[8]. It is clear that all aspects of this issue need to be deeply considered. A very limited number of studies have been performed examining the pathophysiology between PDW and SVG failure since the first description of this relationship by Ege et al.^[2], in 2013. Regarding our study results, platelet activators, such as increased PDW, may initiate or accelerate atherosclerosis. Therefore, it is important to slow down platelet activation and inflammation. Additionally, PDW may not be the only parameter associated with atherosclerotic processes, but the red blood cell distribution width may also play an important role^[9]. The inflammation that is involved in arterial atherosclerosis might contribute to an increased red blood cell distribution^[10]. Although this distribution was not statistically significantly different between the two groups, it approached statistical significance. It can be seen as a subtle contribution.

DM is a well-known cardiovascular risk factor; graft stenosis is more frequent in diabetic patients compared with patients with other risk factors. Surgical treatment is considered to be the standard method for diabetic patients with advanced multivessel diseases^[11]. In addition to this, DM may induce a ‘prothrombotic state’ with enhanced platelet reactivity. Jindal et al^[12]. observed that there is a significant association between PDW and microvascular dysfunction in patients with DM.

Reduced EF was observed in this study to be independently correlated with the rate of vein graft patency, as observed in other studies^[13,14]. This may happen as a result of the need of high coronary blood flow for large myocardial ischemic areas. A normal EF may also indicate ischemic myocardium with a sufficient distal runoff.

An increased time interval after CABS is linked to atherosclerosis of the grafts. Doğan et al.^[15] reported that the age of the SVG is useful for prediction of SVG failure, with the risk increasing over time. We observed similar results in the follow-up of the patients in this study.

The complexity of coronary artery atherosclerosis is determined using the SYNTAX score system. Its validity is well known and well codified^[5]. Considering this information, we found reduced graft patency in patients with high SYNTAX scores. In this context, the SYNTAX score can be seen as a strong predictive factor according to our multivariable model. Atherosclerosis is an ongoing disease process throughout the arterial system^[16]. Correspondingly, patients with CAD may have extensive atherosclerosis in other arterial branches, such as the carotid artery. As a result, these patients may experience adverse events, such as cerebrovascular events, due to the progressive nature of the atherosclerosis. In fact, the rate of transient ischemic attack and/or stroke was higher in patients in the SVG occluded group than in the SVG patent group during the post-surgical period.

The duration of the cardiopulmonary bypass, serum creatinine levels, and mean platelet values are other risk factors for progression of atherosclerosis in the coronary artery in the remote future following coronary artery bypass grafting. However, these factors did not have a significant influence in the multivariable analyses in our study.

CONCLUSION

In conclusion, older age and DM are associated with SVG failure. In addition to these, increased PDW can indicate progression to SVG failure and platelet activation might accelerate disease in the veins, leading to closure. Our results are clinically significant because increased PDW is easily measured and can be used as a simple and valuable marker for prediction of SVG failure.