Abstract
Background Gamma-glutamyl transferase (GGT) level was found to be elevated in plasma of patients with cardiovascular risk factors. The aim of our study was to assess the relationship between serum GGT levels and the occurrence of no-reflow as well as to evaluate the prognostic value of GGT in ST-segment elevation myocardial infarction (STEMI) population.
Methods and Results One hundred sixty-eight consecutive patients with STEMI who underwent percutaneous coronary intervention (PCI) were enrolled in the study. Patients with STEMI were grouped into tertiles according to their admission serum GGT levels. No-reflow after PCI was assessed both angiographically (thrombolysis in myocardial infarction [TIMI] flow and myocardial blush grade) and electrocardiographically (ST resolution). Gamma-glutamyl transferase levels were higher in patients with STEMI compared to the elective PCI group subjects. Patients with angiographically (TIMI flow ⩽2 or TIMI flow 3 with final myocardial bush grade ⩽2 after PCI) and electrocardiographically (ST resolution <30%) detected no-reflow were increased in number across the GGT tertiles. In addition, 1-year mortality rates showed a significant increase across the tertile groups (4% vs 11% vs 23%, P < 0.01). Multivariable logistic regression analysis revealed that GGT levels on admission were a significant predictor of long-term mortality of myocardial blush grade–detected no-reflow phenomenon. High GGT level on admission was a significant predictor for long-term mortality and major adverse cardiac events.
Conclusions In patients with STEMI undergoing primary PCI, high GGT levels at admission were found to be associated with no-reflow phenomenon and increased long-term mortality.
Gamma-glutamyl transferase (GGT) is a plasma membrane enzyme that is used to recycle glutathione for antioxidation.1In a recent study, it was shown that GGT may take a role in the pathogenesis of atherosclerosis and plaque instabilization.2–4Gamma-glutamyl transferase is related with many cardiovascular risk factors, namely, metabolic syndrome, dyslipidemia, diabetes, and hypertension.5–7In addition, serum GGT values were strongly and positively correlated with determinants of oxidative stress such as the levels of C-reactive protein, uric acid, and fibrinogen.8
The primary treatment aim in patients with acute ST-segment elevation myocardial infarction (STEMI) is an effective and timely reperfusion.9Should it be performed in a timely manner by experienced operators, primary percutaneous coronary intervention (PCI) is the preferred strategy for reperfusion in STEMI treatment.10However, the successful restoration of epicardial coronary artery patency does not always end up with improved tissue perfusion, which is a condition known as the no-reflow phenomenon.11The no-reflow phenomenon is encountered in a number of patients with myocardial infarction, ranging from 5% to 50%, according to the methods used for no-reflow assessment and the study population.12The prognostic importance of the no-reflow phenomenon and abnormal myocardial perfusion has been demonstrated by numerous clinical studies, in patients presenting with acute myocardial infarction.13–15Therefore, it can be inferred that prevention and treatment of the no-reflow phenomenon are likely to have an important effect on clinical outcome after primary PCI.
The aim of this study was therefore to determine the relationship between GGT levels and the occurrence of angiographically and electrocardiographically detected no-reflow phenomenon in patients with STEMI. A secondary objective was to evaluate the prognostic value of GGT in the STEMI population reperfused by primary PCI.
MATERIALS AND METHODS
Study Patients
One hundred eighty-one consecutive patients with a definite diagnosis of STEMI within 12 hours from the onset of symptoms were primarily enrolled in the study. Patients with cardiogenic shock within the first 24 hours were also included. ST-segment elevation myocardial infarction was diagnosed in the presence of the 2 following criteria: persistent anginal pain for 20 minutes or more and ST-segment elevation of more than 1 mm in 2 or more standard leads or 2 mm or more in 2 or more contiguous precordial leads or the presence of new left bundle branch block.16Percutaneous coronary intervention was preferred as the reperfusion strategy for all patients because of its availability in our center and superiority to fibrinolytic therapy.17The thrombolysis in myocardial infarction (TIMI) risk score was calculated for all patients.18
Patients with culprit lesion in left main coronary artery or left main stenosis of more than 50%, previous coronary artery bypass surgery, those with end-stage renal disease (creatinine clearance <15 mL/min), hematological disorders, active hepatobiliary disease, alcohol intake, active infections, neoplastic diseases, hepatic steatosis or fibrosis confirmed by ultrasonography, recent major surgical procedure or trauma, and patients with insufficient data were excluded from the study. Thus, the population of the analysis consisted of 168 patients with STEMI. Informed consent was obtained from all participating patients. Permission for the study was obtained by the local ethics committee.
Measurements
Venous blood samples were obtained on admission in the coronary care unit or emergency department before PCI. The activity of GGT was measured by using a Modular P-800 autoanalyzer with original kits (Roche, Basel, Switzerland). High-sensitivity C-reactive protein (hs-CRP) levels were measured by immunonephelometric method (Immage Immunochemistry System, Beckman Coulter, Inc, Fullerton, CA). Creatinine kinase-myoclonal band measurements were made with Roche Hitachi Modular P800 (Tokyo, Japan) by spectrophotometric method. Troponin I was measured Siemens Immulite2000 Immunoassay System (Germany). Other biochemical parameters including lipid profile were measured by using commercially available methods and kits.
Angiographic Analysis
Coronary angiography was performed with the Judkins technique using standard projections and recorded in digital media for quantitative analysis (Dicom-viewer; MedCom GmbH, Darmstadt, Germany). Percutaneous coronary intervention was performed according to standard clinical practice using a nonionic, low-osmolality contrast agent, iopromide (Ultravist 370, Schering, Berlin, Germany). Additional thrombectomy was performed only in a limited number of patients. All patients received 300 mg of acetylsalicylic acid and 600-mg loading dose of clopidogrel before intervention and unfractioned heparin during PCI on a routine basis. Only bare metal stents were used for all patients. The operator acted on his own initiative to start/use glycoprotein IIb/IIIa receptor blockers. Clopidogrel at a dose of 75 mg/d was prescribed to each patient for of at least 1 month if stent was implanted.
Digital angiograms were then analyzed by 2 independent experienced interventional cardiologists blinded to the data. Coronary blood flow patterns before and after primary PCI were assessed according to the TIMI flow grade (TFG), using grades 0, 1, 2, and 3.19Myocardial blush grade (MBG) was assessed by the technique defined by Van’t Hof et al.20According to Van’t Hof et al.,20MBG = 0 if there is no contrast density, MBG = 1 with minimal contrast density, MBG = 2 with moderate contrast density but less than normal, and MBG = 3 if the contrast density is normal. We defined angiographic no-reflow phenomenon as a coronary TIMI flow 2 or less after vessel reperfused or a TIMI 3 with a final MBG 2 or less according to previous studies.21Intraobserver and interobserver variabilities for TIMI 0 to 1 were 6% and 7%, respectively; for TIMI 2, the variability values were 4% and 4%, respectively, whereas for TIMI 3, the values were 1%; and for the MBG readings, the values were 5%, and 7% respectively, which were obtained from a random sample of 100 patients.
Electrocardiographic and Echocardiographic Analysis
The percentage of ST-segment resolution (STR) (90 minutes after PCI) was calculated as described previously.22The patients were divided into 3 groups: (1) complete (≥70%), (2) partial (<70%-30%), and (3) no (<30%) STR.22A 2-dimensional echocardiogram was performed for the evaluation of left ventricle ejection fraction (LVEF) by using the modified Simpson technique. The analyses were carried out by 2 observers blinded to the clinical and angiographic data. The intraobserver and interobserver variabilities for electrocardiographic analysis were 5%, and 7%, respectively, and the intraobserver and interobserver variabilities for echocardiographic analysis were 6% and 7%, respectively, derived from a random sample of 100 patients.
Clinical Follow-Up and Study End Points
Clinical follow-up data were obtained from outpatient examination or by telephone contact at 12 months (interquartile range [IQR], 9-15 months) after PCI. Primary end points were long-term all-cause death, and the occurrence of major adverse cardiovascular events (MACE) defined as a composite of death, nonfatal reinfarction, target vessel revascularization (TVR), or new-onset congestive heart failure during hospitalization and at 1-year clinical follow-up. Inhospital reinfarction was defined as recurrent chest pain lasting more than 30 minutes, associated with new Q waves or recurrent ST-segment elevation 1 mm or more in leads and a re-elevation of creatine kinase-myoclonal band to at least twice the upper limit of the reference range and/or more than 50% higher than the previous value after index procedure.23Data of reinfarction and TVR that occurred after discharge were obtained from outpatient examinations together with telephone contacts. New-onset heart failure was defined as any heart failure with symptoms of New York Heart Association class III to IV occurring more than 24 hours after the index event. Target vessel revascularization was described as percutaneous intervention or surgical bypass grafting of any segment of the target vessel after the primary procedure. The target vessel was defined as the whole major coronary artery proximal and distal to the target lesion including all branches and the target lesion itself.24
Statistical Analysis
Analyses were performed using SPSS software (version 15.0, SPSS, Chicago, IL). Continuous data were presented as median ± IQR or mean ± SD. The study population was grouped into tertiles according to the admission serum GGT levels. Comparisons of multiple mean values were carried out with Kruskal-Wallis tests or analysis of variance as appropriate. To test the distribution pattern, the Kolmogorov-Smirnov test was used. Categorical variables were summarized as percentages and compared with the χ2 test. The Spearman correlation coefficient was computed to examine the association between 2 continuous variables. Multivariable logistic regression analysis was applied to identify whether GGT on admission was independently associated with MBG-detected no-reflow phenomenon. At this juncture, all variables showing a significant association with abnormal perfusion with MBG-detected no-reflow phenomenon at univariable analysis were included in the model. For univariable analyses, all variables in Table 1 were tested.
Multivariate Cox regression analyses were used for the prediction of mortality and MACE at 1 year in models including variables that were found to be predictive by univariate analysis. Univariate correlations with P < 0.10 were encountered into multivariate models. Analysis was performed as 2 models; in the first model, admission GGT level was assumed as a continuous variable; whereas in the second model, admission GGT level was assumed as a categorical variable. The Kaplan-Meier method was used to illustrate the timing of events during long-term follow-up in relation to GGT level on admission, and statistical assessment was performed using the log-rank test. An exploratory evaluation of additional cut-points was performed using the receiver-operating characteristics curve analysis. P < 0.05 was considered statistically significant.
RESULTS
Baseline Characteristics
This prospective study included 168 patients with STEMI. Patients were subgrouped into tertiles according to the serum GGT activity: less than 23 U/L in the first tertile (T1), 23 to 42 U/L in the second tertile (T2), and more than 42 U/L in the third tertile (T3). Demographic and clinical characteristics are shown in Table 1. There were no significant differences between the groups regarding baseline characteristics (age, sex, risk factors, etc), TIMI risk score, and reperfusion times; but serum hs-CRP levels increased across the tertiles. ST-segment resolution (≥70%) and LVEF was decreased across the tertile groups (Table 1).
GGT and No-Reflow Phenomenon Assessed by ST-Segment Resolution and TIMI Flow/MBG
Of the 168 patients, 32 patients (19%) had no ST-resolution (<30%), which was consistent with electrocardiographically detected no-reflow phenomenon. Patients with no ST resolution had significantly higher GGT values on admission in comparison with patients with intermediate or complete ST resolution (31%–100%) (42 [IQR, 26–58] vs 24 [IQR, 20–43], P < 0.001; Fig. 1). There were no significant differences between the groups regarding TIMI flow zero before PCI. After PCI, a total of 40 patients (24%) had TIMI flow 2 or less and 148 patients (76%) had TIMI flow 3. Angiographic no-reflow phenomenon (TIMI flow grade 2 or less after PCI or TIMI flow grade 3 with a final MBG 2 or less) was observed in 77 patients (46%), with a trend increase across the tertile groups (Table 2). Patients with angiographically detected no-reflow after PCI had also significantly higher GGT levels on admission (43 [IQR, 23–58] vs 23 [IQR, 20–34], P < 0.001; Fig. 1).
Predictors of MBG-Detected No-Reflow
In a multivariable logistic regression model adjusted for significant variables in univariable regression analysis (GGT levels on admission, LVEF, TIMI risk score, significant STR, and TIMI flow before PCI) using MBG-detected no-reflow as the dependent variable, the only significant predictors of no-reflow phenomenon were significant STR (odds ratio [OR], 0.80; 95% confidence interval [CI], 0.73–0.90), TIMI risk score (OR, 1.25; 95% CI, 1.12–1.42), and GGT level on admission (OR, 1.08; 95% CI, 1.04–1.19). Admission troponin values did not predict no-reflow (OR, 1.017; 95% CI, 1.04–1.19; P = 0.22)
Clinical Outcomes
Patients who died during follow-up had significantly higher median levels of GGT when compared with patients who survived (50 [IQR, 43–82] U/L vs 25 [IQR, 20–44] U/L, P < 0.01). One-year mortality frequencies were significantly increased across the tertile groups (Table 2).
In the third tertile group, 3 patients (5%) were observed to develop reinfarction, 4 patients (7%) were observed to develop TVR, and 2 patients (4%) were observed to develop congestive heart failure. Nonfatal reinfarctions, TVR, and congestive heart failure occurred in 2 patients (4%), 3 patients (5%), and 2 patients (4%) in tertile 2 on admission group, respectively. In the first tertile group, 2 patients (4%) were observed to develop reinfarction, 2 patients (4%) were observed to develop TVR, and 1 patient (2%) was observed to develop congestive heart failure. As a result, in long-term follow-up, MACE frequencies were observed to increase significantly across the tertile groups (Table 2).
When we assumed GGT as a continuous variable in the multiple Cox regression analysis (model 1), TIMI risk score, high GGT level, angiographic no-reflow, hs-CRP, and LVEF emerged as independent predictors of mortality. In the same way, when we assumed GGT as a categorical variable (model 2), TIMI risk score, angiographic no-reflow, and GGT of 42 U/L or greater level emerged as independent predictors of mortality (Table 3).
Significant predictors of MACE in the Cox regression analysis are shown in Table 4. On multivariable analysis (model 1), the only significant predictors of MACE were TIMI risk score, prior MI, angiographic no-reflow, and increased GGT levels. On multivariable analysis model 2, the only significant predictors of MACE were TIMI risk score, angiographic no-reflow, and GGT of 42 U/L or greater (Table 4).
The receiver-operating characteristic curve analysis further revealed that GGT levels on admission were a strong indicator of mortality with an area under the curve of 0.82 (95% CI, 0.74-0.90) compared with other strong established marker of survival after STEMI (TIMI risk score area under the curve, 0.89 [95% CI, 0.80-0.98]). The optimal threshold of GGT that maximized the combined specificity and sensitivity to predict mortality was 42 U/L (Fig. 2). Sensitivity, specificity, positive predictive value, and negative predictive value to identify death were 82%, 74%, 30%, and 97%, respectively, with GGT greater than 42 U/L. Figures 3 and 4 show the Kaplan-Meier curves for 1-year mortality rate and MACE in patients with high GGT level (≥42 U/L) (T3) versus those with low GGT level (<42 U/L) (T1 and T2). By Spearman correlation analysis, GGT was positively related to hs-CRP (r = 0.40; P < 0.01).
DISCUSSION
The cardinal findings of the present study indicated that: (1) Angiographically and electrocardiographically detected no-reflow increased across the GGT level tertiles, and (2) elevated serum GGT activity in patients with STEMI undergoing primary PCI may be correlated with no-reflow phenomenon and increased long-term mortality.
Epidemiological studies have demonstrated that GGT is related with mortality from all causes: MI, stroke, and cardiac death. This association was independent from hepatic disease, alcohol intake, and cardiovascular risk factors.25–28It has been found that high GGT values were positively associated with fatal events in chronic forms of coronary artery disease, congestive heart failure, and ischemic and hemorrhagic stroke. In a recent report, it is shown that serum GGT activity was independent predictor for early mortality in a group of STEMI patients.29,30
Although the exact pathophysiological mechanisms by which elevated admission serum GGT activity increase the risk for poor myocardial perfusion after primary PCI are not clearly clarified, 2 possible mechanisms may be oxidative stress and inflammation. Gamma-glutamyl transferase is associated with oxidative stress pathways in several organ systems, localizes to atheromatous plaques containing oxidized low-density lipoprotein, and is a proinflammatory molecule, which leads us to further implicate this protein in atherogenesis.31–33Gamma-glutamyl transferase can promote low-density lipoprotein oxidation by hydrolyzing glutathione,34the possibility exists that circulating GGT may participate in the pathogenesis of cardiovascular atherosclerotic disease. Gamma-glutamyl transferase activity increases to compensate the increased oxidative stimulants by increasing the glutathione synthesis during the atherosclerotic process. In animal studies, it was shown that glutathione peroxidase-1 has a significant protective effect from ischemic reperfusion injury in myocardium.35,36Furthermore, in a clinical trial, vitamin C, vitamin E, and glutathione peroxidase levels were demonstrated to be significantly lower in a group with no-reflow than in a group with reflow before PCI in patients with STEMI, whereas no significant difference in the levels of beta-carotene, superoxide dismutase, or catalase were observed.37Despite the fact that the no-reflow phenomenon can initially be demonstrated by analysis of TIMI flow grade,19the integration of MBG and STR has been shown to improve the risk stratification of patients.38,39In our study, no difference in TFG-evaluated epicardial coronary flow was observed among different serum GGT groups; however, a significant difference was detected when evaluated by MBG, which had been much more sensitive and useful than TFGs and had been associated better with impaired microvascular flow. Additionally, we observed that the rate of impaired angiographic reperfusion increased across the tertiles comprising patients with different GGT levels. As mentioned earlier, this increase in GGT activity may be associated with the increased oxidative stress, which is also responsible for the abnormal macrovascular and microvascular flow.
One other explanatory mechanism for the association of increased GGT activity and the poor myocardial perfusion may be inflammation. It was reported that hs-CRP measured at either presentation or hospital discharge was proposed to have prognostic value in patients with acute coronary syndrome.40,41Another clinical study demonstrated that admission hs-CRP levels may predict the efficacy of reperfusion in patients with acute myocardial infarction.42In our study, serum GGT activity and serum hs-CRP levels had been found to be independent risk factors for 1-year mortality in patients with STEMI undergoing primary PCI. In addition, serum GGT activity on admission showed a significant positive correlation with serum hs-CRP levels. In previous studies, GGT and CRP or other inflammatory parameters were found to be associated, suggesting that this enzyme may be related to subclinical inflammation and cellular stress.8,31,43,44Oxidative processes might have an implication in chronic inflammation may involve an oxidative process45; it has been postulated that elevation in GGT might precede an elevation in CRP, and the related oxidative stress would give rise to a subsequent inflammatory response.31
Although it is a known fact that high GGT activity is associated with poor long-term prognosis in chronic coronary artery disease and particularly in patients with STEMI, in this very study, we found that high GGT activity was related with abnormal myocardial flow and no-reflow phenomenon, which in turn, may contribute to poor outcomes. In a recent study, Akpek et al.30reported that serum GGT activity was associated with inhospital MACE in patients with STEMI undergoing primary PCI but not with impaired coronary flow. They emphasized the role of oxidative stress and its link with increased GGT activity. They also highlighted the possible association of the GGT activity and hyperactivity of platelets. However, why they did not find an association with GGT activity and impaired coronary flow may be related to the method they used to estimate impaired flow. The method they used to detect impaired coronary flow was the TFG. No-reflow phenomenon can initially be demonstrated by analysis of TFG.31Indeed, in 5% to 10% of patients with TIMI flow grade 0 to 2, it is predictably associated with the no-reflow phenomenon. On the contrary, however, no re-flow also occurs in a sizeable proportion of patients with apparently successful large epicardial vessel reopening resulting in TFG 3. Thus, the sensitivity of TIMI flow assessment in the detection of no-reflow phenomenon is rather low. At the time of primary PCI, no-reflow can be realized more effectively by assessing MBG, which had been much more sensitive and useful than TFGs and had been associated better with impaired microvascular flow. Besides, MBG was an independent risk predictor of 1-year mortality in our study. Additionally, we found that admission troponin values were not predictive of no-reflow.
The primary limitations of the present study are its reflection of a single-center experience and the limited number of patients. However, one strength of the study is that it is composed of a homogeneous consecutive unselected group of patients that is directly relevant to most patients undergoing STEMI in the general population. Another limitation of this study is that only admission GGT levels were evaluated. Gamma-glutamyl transferase levels were not evaluated after the acute phase of the myocardial infarction. Management of the patients was per protocol of the unit identical in all patients.
CONCLUSION
In conclusion, increased serum GGT activity is associated with no-reflow phenomenon in patients with STEMI undergoing primary PCI, which may contribute to poor long-term prognosis. Therefore, admission serum GGT activity detection may be helpful in identifying a subgroup of patients at a greater risk for no-reflow and worse long-term prognosis. To improve the cardiovascular outcomes, we should expand the “inexpensive and practical tools” paradigm such as admission GGT levels in the prediction of no-reflow phenomenon with further large-scale and randomized study data.
ACKNOWLEDGMENT
The authors thank Selcuk Kanat and Ahmet Isleyen for their technical assistance.