Platelet aggregates appear to have a pathogenic role in the no-reflow phenomenon, which is associated with impaired clinical outcome in patients with ST-segment elevation myocardial infarction (STEMI). Melatonin, a hormone that plays a major role in biological circadian rhythms, is present in human platelets. Lowered circulating melatonin levels predict poor outcome in patients with STEMI undergoing primary percutaneous coronary intervention (PPCI). We investigated whether intraplatelet melatonin levels correlate with angiographic no-reflow after PPCI in patients with STEMI. We studied 180 consecutive patients with a first STEMI who underwent PPCI within 6 hours from onset of symptoms. Intraplatelet melatonin levels were measured in platelet-rich plasma using an enzymatic immunoassay method. After PPCI, angiographic no-reflow (defined as Thrombolysis In Myocardial Infarction grade <2 flow) was observed in 63 patients (35%). Patients with angiographic no-reflow had lower intraplatelet melatonin levels compared to patients without no-reflow (12.32 ± 3.64 vs 18.62 ± 3.88 ng/100,000 platelets, p <0.0001). After adjusting by potential confounders, binary logistic regression analysis showed that intraplatelet melatonin levels were the only significant predictor of angiographic no-reflow (odds ratio 1.58, 95% confidence interval 1.37 to 1.82, p <0.0001). In conclusion, low intraplatelet melatonin concentration predicts angiographic no-reflow after PPCI in patients with STEMI.
Melatonin, a hormone associated with biological circadian rhythms and an antioxidant substance in its own right, might be implicated in the no-reflow process. Evidence gathered in the previous 15 years suggests a circadian regulatory influence of melatonin on the cardiovascular system. Moreover, other cardiovascular actions been attributed to melatonin, e.g., patients with coronary artery disease and low melatonin production rates have been reported to be at a higher risk for myocardial infarction and sudden cardiac death. Melatonin has been suggested to have a protective effect on the cardiovascular system mediated by melatonin receptor-independent actions and mainly by its radical scavenger and anti-inflammatory functions. Melatonin synthesis is not restricted to the pineal gland but also takes place in megakaryocytes and platelets. Platelets have high-affinity binding sites for melatonin and an antiaggregant action has been demonstrated for this hormone. In an in vitro study, Martinez-Campa et al showed that melatonin inhibits cyclo-oxygenase 1 and 2 mRNA expressions in MCF-7 cells. Thus, increasing evidence suggests a modulatory role of melatonin on platelet function. In the present study we therefore sought to investigate whether lower preprimary percutaneous coronary intervention (PPCI) intraplatelet melatonin levels are associated with angiographic no-reflow in patients with ST-segment elevation myocardial infarction (STEMI).
Methods
Two hundred two consecutive patients with STEMI without previous coronary artery disease underwent PPCI within 6 hours of pain onset in Hospital Universitario de Canarias institution and were enrolled in the study. Diagnosis of STEMI was established by the presence of prolonged chest pain (>30 minutes) with ST-segment elevation >0.2 mV in ≥2 adjacent leads on standard electrocardiogram. Twenty-two patients were excluded from analysis if time from symptom onset was >6 hours (n = 12), there was documentation of previous coronary artery disease on further assessment during hospital admission (n = 8), and had missing blood samples for intraplatelet melatonin measurement (n = 2). Thus, 180 patients were included in the study. Pharmacologic treatment on admission to the emergency room included, in every case, oral administration of aspirin (300 mg) and clopidogrel (600 mg), 20 to 30 minutes before peripheral blood sampling, which was carried out when a patient arrived to the catheterization laboratory for PPCI. Standard therapies after PPCI included aspirin, clopidogrel, β blockers, statins, and angiotensin-converting-enzyme inhibitors or angiotensin II receptor blockers, according to current guidelines. The research protocol was approved by the ethics committee of Hospital Universitario de Canarias institution and all patients gave written informed consent for inclusion in the study.
The right femoral approach was used, with a 6Fr guiding catheter, and a bolus of heparin 5,000 IU was administered to all patients. After PPCI intracoronary nitrates were given to ensure maximal epicardial vasodilation to determine the size and length of the stent and facilitate stent placement. Intracoronary or systemic abciximab (bolus) was given, followed by a 12-hour continuous infusion, and a device for thrombus aspiration was used at the operator’s discretion. The status of the target coronary lesion was assessed before and after PPCI using standardized projections. These allowed optimal evaluation of the Thrombolysis In Myocardial Infarction (TIMI) flow. Angiographic assessment of stenosis severity and TIMI blood flow were performed off-line by 2 independent angiographers who were blinded to intraplatelet melatonin levels and electrocardiographic (ECG) data. In addition, calculations were carried out on 60 randomly selected angiograms to determine interobserver and intraobserver variabilities as assessed by weighted kappa statistics. For TIMI flow grade, kappa values were 0.91 (95% confidence interval 0.78 to 0.99) for interobserver measurements and 0.85 (95% confidence interval 0.77 to 1.03) for intraobserver measurements. Disagreement was resolved by consensus. Angiographic no-reflow after PPCI was defined as TIMI grade <2 flow during stent implantation without evidence of dissection, stenosis, or vasospasm. ST-segment elevation was analyzed by a single investigator with lens-intensified calipers to the nearest 0.025 mV 20 ms after the end of the QRS complex with the PR segment as reference baseline from leads I, aVL, and V 1 through V 6 for anterior infarction and leads II, III, aVF, V 5 , and V 6 for inferior infarction. In every patient, 12-lead electrocardiograms were recorded before PPCI and 90 minutes after the procedure. ST-segment resolution, expressed as percent change, was calculated using the following formula: (Σ ST elevation before PPCI − Σ ST elevation after PPCI)/Σ ST elevation before PPCI. Lack of ST resolution was defined as a decrease ≤50% of the measured ST-segment elevation at 90 minutes compared to basal electrocardiogram.
Peripheral venous blood samples were obtained in all patients at hospital admission for intraplatelet melatonin assessment and immediately after accessing the femoral artery before heparin and abciximab administration. For platelet-rich plasma, 10 ml of venous blood was collected into siliconized Vacutainer tubes containing ethylenediaminetetra-acetic acid (BD Vacutainer, BD Co, Franklin Lakes, New Jersey). Platelet-rich plasma was prepared by differential centrifugation (900 × g for 70 seconds and 10,000 × g for 5 minutes, respectively) at room temperature. The pellet of platelets was washed twice in cold saline solution and centrifuged again. The pellet was resuspended in saline solution and counted using a Coulter Haematology Analyzer (Coulter Electronic, Teltow, Germany). To obtain the homogenate, platelets were destroyed by sonication (100 W for 30 seconds; Ultrasonic Laboratory Devices, Hamburg, Germany). The intraplatelet melatonin content was measured by an enzyme-immunoassay method using a commercial kit purchased from Immuno Biological Laboratories (IBL Hamburg, GmbH, Hamburg, Germany). In this enzyme-linked immunosorbent assay, the lowest detection limit was 2.58 pg/ml. Coefficients of variation were 7.53% and 11% for intra- and interassay variabilities, respectively. Serum C-reactive protein (CRP) was measured by an ultrasensitive enzyme-linked immunosorbent assay kit (DRG Instruments, GmbH, Marburg, Germany). In this enzyme-linked immunosorbent assay, the lowest detection limit of CRP was 0.010 mg/L. Coefficients of variation were 5.12% and 11% for intra- and interassay variabilities, respectively. Serum troponin I was determined immunoenzymatically using a technique based on sandwich enzyme-linked immunosorbent assay (Boehringer Mannheim, Mannheim, Germany). Coefficients of variation were 2.2% and 5.9% for intra- and interassay variabilities, respectively. Blood samples for troponin I assessment were taken every 8 hours during the first day and every 24 hours in the next 3 days according to hospital protocol. In all patients, left ventricular ejection fraction was obtained using a commercially available system (Philips Medical Systems iE33, Boston, Massachusetts). Echocardiograms were obtained by experienced ultrasonographers after the PPCI procedure and were blinded to intraplatelet melatonin, CRP measurements, and clinical data. Ejection fractions were calculated using the modified Simpson rule. Coefficients of variation in the evaluation of ejection fractions were 3.5% and 3.9% for intra- and interobserver variabilities, respectively.
Continuous variables are reported as mean ± SD, and categorical variables as frequencies and percentages. Comparisons between groups were carried out by Mann-Whitney U test for continuous variables and by Fisher’s exact test or chi-square test for discrete variables. Multivariable binary logistic regression analysis was applied to identify whether intraplatelet melatonin, included as a continuous variable, was independently associated with angiographic no-reflow. Backward stepwise selection was used in the multivariate model to derive the final model for which significance levels of 0.1 and 0.05 were chosen to exclude and include terms, respectively. We included in the model variables known from the literature to be associated with angiographic no-reflow. We followed the recommendations of Peduzzi et al that “in case of a dichotomy dependent variable should have least 10 cases for each of the two values to obtain reliable estimates” and we included a maximum of 6 variables in the analysis. The main independent variable was intraplatelet melatonin, the dependent variable was angiographic flow, and the controlled variables were culprit left anterior descending coronary artery, time from chest pain onset to PPCI, thrombus-aspirating device usage, troponin I peak, and left ventricular ejection fraction. The magnitude of the effect is expressed as odds ratio and its 95% confidence interval. A p value <0.05 was required for statistical significance. All tests were 2-tailed and analyses were performed using SPSS 17.0 (SPSS, Inc., Chicago, Illinois).
Results
Baseline clinical characteristics are listed in Table 1 . After PPCI, incidence of angiographic no-reflow was 35% and that of ECG no-reflow was 24%. Our population included 63% of patients with TIMI grade 0 flow on admission. Abciximab was used in 100% of cases, whereas thrombus aspiration was performed in nearly 1/3 of cases. Patients with angiographic no-reflow had lower levels of intraplatelet melatonin compared to patients without angiographic no-reflow ( Figure 1 , Table 1 ). Regarding ECG findings, intraplatelet melatonin levels were lower in patients who had ECG no-reflow compared to those who did not, but without reaching statistical significance (15.71 ± 5.44 vs 16.68 ± 4.62 ng/100,000 platelets, respectively, p = 0.32). Incidence of angiographic no-reflow significantly decreased with higher tertiles of intraplatelet melatonin (p <0.0001; Figure 2 ).
| Variable | All Patients (n = 180) | Angiographic No-Reflow | p Value | |
|---|---|---|---|---|
| Yes (n = 63) | No (n = 117) | |||
| Age (years) | 61 ± 13 | 62 ± 12 | 60 ± 13 | 0.58 |
| Men | 142 (78%) | 51 (81%) | 91 (77%) | 0.61 |
| Hypertension (>140/90 mmHg) | 125 (69%) | 45 (71%) | 80 (68%) | 0.67 |
| Hypercholesterolemia (>5.17 mmol/L) | 70 (38%) | 25 (39%) | 45 (38%) | 0.87 |
| Smokers | 90 (50%) | 27 (42%) | 63 (53%) | 0.16 |
| Diabetes mellitus | 54 (30%) | 19 (30%) | 35 (29%) | 0.97 |
| Pain to balloon time (min) | 220.98 ± 74.60 | 220.65 ± 75.20 | 221.17 ± 74.60 | 0.97 |
| Culprit left anterior descending coronary artery | 100 (55%) | 39 (61%) | 61 (52%) | 0.20 |
| Thrombus-aspirating device | 50 (27%) | 19 (30%) | 31 (26%) | 0.60 |
| Abciximab usage | 180 (100%) | 63 (100%) | 117 (100%) | 1 |
| Left ventricular ejection fraction (%) | 55 ± 10 | 53 ± 10 | 56 ± 10 | 0.02 |
| Thrombolysis In Myocardial Infarction grade 0 flow | 115 (63%) | 42 (66%) | 73 (62%) | 0.92 |
| Troponin I peak (ng/ml) | 80.83 ± 16.06 | 84.04 ± 14.08 | 79.10 ± 16.83 | 0.08 |
| White blood cell count (×10 9 /L) | 10.4 ± 3.2 | 10.3 ± 3.1 | 10.5 ± 3.3 | 0.76 |
| Platelet count (×10 9 /L) | 243 ± 69 | 244 ± 63 | 243 ± 73 | 0.74 |
| C-reactive protein (mg/L) | 15.27 ± 0.85 | 5.43 ± 0.90 | 5.19 ± 0.81 | 0.09 |
| Melatonin (ng/10 5 platelets) | 16.41 ± 4.84 | 12.32 ± 3.64 | 18.62 ± 3.88 | <0.0001 |
In baseline analysis, intraplatelet melatonin levels (p <0.0001) and left ventricular ejection fraction (p = 0.02) were significantly associated with angiographic no-reflow, whereas troponin I peak and CRP were of borderline statistical significance ( Table 1 ). Table 2 presents simple logistic regression analyses using angiographic no-reflow as a dependent variable and intraplatelet melatonin levels, culprit left anterior descending artery, time to PPCI, thrombus-aspirating device usage, troponin I peak, and left ventricular ejection fraction as independent variables. After adjusting by possible confounders, multivariate logistic regression analysis showed intraplatelet melatonin levels (p <0.0001) to be the only significant predictor of angiographic no-reflow, whereas troponin I and time from chest pain onset to PPCI were of borderline statistical significance ( Table 3 ).
