Evaluation of Serum Serotonin as a Biomarker for Myocardial Infarction and Ischemia / Reperfusion Injury

: Background: Activated platelets release serotonin during acute myocardial infarction (AMI), aggravating myocardial damage and ischemia / reperfusion (I / R) injury. However, serum serotonin and its potential role as a biomarker for myocardial infarction and I / R injury have not been studied so far. Methods: In this investigator-initiated pilot study, we examined 38 patients with ST-segment myocardial infarction (STEMI). We determined serum serotonin levels prior to percutaneous coronary intervention and 8, 16, and 24 h afterwards. We studied whether serum serotonin was associated with I / R injury assessed by ECG analysis and by analysis of TIMI myocardial perfusion grade (TMP) and myocardial blush grade (MGB). Serum serotonin levels were compared to an age-matched control group consisting of patients admitted to the emergency department for any other reason than STEMI. Results: Serum serotonin levels were not elevated in the myocardial infarction group compared to the control cohort and they did not show any timeline kinetics after STEMI. They were not associated with the severity of coronary artery disease, the outcome of coronary angiography, the extent of I / R injury, or the degree of heart failure. Conclusions: Serum serotonin is not suitable as a biomarker after myocardial infarction and in the assessment of I / R injury.


Introduction
Despite therapeutic advances in recent decades, coronary artery disease (CAD) and its sequela acute myocardial infarction (AMI) are still a major cause of morbidity and mortality worldwide [1,2]. In AMI, early recognition and initiation of reperfusion therapy are crucial to limit myocardial ischemia

Patient Population
This investigator-initiated pilot single-center study was conducted at the University Heart Center, Freiburg between March 2017 and May 2018. The protocol of this study conforms to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the institutional ethical committee of the University of Freiburg (EK 137/16). Written informed consent was obtained from all participants prior to inclusion.
Patients admitted to our department due to ST-segment myocardial infarction (STEMI) with the objective of immediate restoration of blood flow by coronary intervention were screened for eligibility. Exclusion criteria were sepsis, life-limiting oncologic or hematological diseases, or the inability to provide informed consent. The final study population consisted of 38 patients. None of the patients reported the intake of serotonin reuptake inhibitors (SSRIs) or any other psychiatric medication.
Age-matched patients admitted to the department of emergency medicine of the University Medical Center-University of Freiburg for any other reason than STEMI served as the control group. The collection of serum from these patients was approved by the institutional ethical committee of the University of Freiburg (EK 153/20). Patients were recruited between March 2020 and June 2020, and again, written informed consent was obtained from all participants prior to inclusion. Only patients without intake of antiplatelet therapy such as, e.g., ASS, clopidogrel, ticagrelor, or prasugrel, were included in the analysis for this manuscript.

Primary Percutaneous Coronary Intervention (PCI)
PCI in the STEMI group was performed according to the European Society of Cardiology (ESC) guidelines [27]. Prior to intervention, unfractioned heparin (5000IE) was administered. Moreover, all STEMI patients received aspirin (minimum of 250 mg) and either prasugrel 60 mg or ticagrelor 180 mg. None of our patients received sarpogrelate, a 5HT2A antagonist. We analyzed the thrombolysis in myocardial infarction (TIMI) grade prior and post coronary angiography to assess the success of the PCI [28].

Blood Collection
In the STEMI group, venous blood was collected into serum tubes (Sarstedt, Nümbrecht, Germany) prior to coronary angiography and 8, 16 and 24 h after intervention. After collection, the blood was incubated at room temperature for 20 min and then centrifuged (1500× g, 20 • C). The resulting supernatant (serum) was aliquoted and stored at −20 • C.
Serum samples from the control group were obtained immediately after admission to the emergency department.

Patient Characteristics/Laboratory Findings
Patient characteristics were obtained from the electronic health record. Laboratory values such as C-reactive protein (CRP), serum creatinine, serum urea (BUN), creatine kinase (CK), myoglobin, brain natriuretic peptide (pro-BNP), and troponin were assessed and obtained from the electronic health record.

Ischemia/Reperfusion Injury
In the STEMI group, ECGs from 17 patients were analyzed before and after (mean 41 ± 11 min) PCI. A decrease in ST-elevation by a minimum of 70% was set as the cut-off for a patient to be considered adequately reperfused [29]. Patients could then be subdivided into adequately reperfused or reperfusion injury groups according to this ECG criterion [29][30][31][32]. For the assessment of I/R injury, we did not only use the surrogate parameter of ST-elevation resolution, but also determined the visually direct markers, TIMI myocardial perfusion grade (TMP) and the mycoardial blush grade (MGB) [33,34]. Analysis of coronary angiography was conducted by a blinded investigator.

Determination of Serum Serotonin Levels
Serotonin was analyzed in the serum samples using an enzymatic immunoassay kit (Serotonin ELISA FastTrack , LDN, Nordhorn, Germany). The ELISA was performed on samples in duplicate following the manufacturer's instructions. All samples and standards were analyzed on a microplate reader set at 450 nm.
According to the data sheet, the detection range of the ELISA was 15-2500 ng/mL and the reference range for serum serotonin was 70-270 ng/mL.
Statistically significant outliers were excluded using Gubb's test. Continuous variables were tested for normal distribution by using the Shapiro-Wilk test. Data are presented as mean ± standard deviation if found to follow a Gaussian distribution, otherwise as median with interquartile range.
Variables following a Gaussian distribution were compared using a Student's t-test or ANOVA with Bonferroni correction, non-normally distributed continuous values were compared by using a Mann-Whitney U test or Kruskal-Wallis test with Dunn's correction. For the analysis of repeated measurements (all data in this case were non-parametric), the Friedman test with Bonferroni correction was applied.
Correlation analysis was performed using the Pearson correlation test for parametric data or by a Spearman test for non-parametric data.
A p-value less than 0.05 was considered statistically significant for all analyses except for the analysis of correlation. This analysis required correction by Bonferroni's method. This set the new statistically significant p-value at 0.0022.

Patient Characteristics
The average age across the STEMI group was 60.1 ± 12.9 years and 61.6 ± 21.3 years in the control group (p = 0.7002). The majority of the STEMI patients were male (76.3%), compared to 46% in the control cohort, however, the difference was not statistically significant (p = 0.818). Serum creatinine, urea (BUN), or CRP did not differ between both groups (Table 1). The STEMI patients mostly featured a pronounced cardiovascular risk profile. Of them, 60.5% suffered from arterial hypertension, 50% from hypercholesterolemia, and 26.3% from diabetes mellitus. Furthermore, 39.5% were current smokers and 15.8% had quit smoking in recent years. The culprit lesion in most of our STEMI patients was either RCA (44.7%) or LAD (47.4%). The median left ventricular ejection fraction (LVEF) after myocardial infarction was 45.0% (Table 2). Table 2. Coronary artery disease (CAD) characteristics in the ST-elevation myocardial infarction (STEMI) cohort. a Presented as median ± interquartile range, b presented as mean ± standard deviation, c number of patients (with percentage based on the number of patients with a non-missing value for that characteristic. Hypercholesterolemia was defined as total cholesterol >200 mg/dL or on medication. For creatine kinase (CK), creatine kinase myocardial band (CK-MB), myoglobin, troponin, and brain natriuretic peptide (BNP), the peak values within the first 24 h after myocardial infarction are listed.

Kinetics of Serum Serotonin
Serum serotonin levels ranged between 21.3 and 325.3 ng/mL, with a median of 114.2 ng/mL (IQR: 71.9-135.3). Median and interquartile range were within the reference range for serum serotonin levels (70-270 ng/mL).
The first samples were taken prior to coronary angiography (t = 0) and subsequently 8, 16 and 24 h later. We did not see any significant increase or decrease in serum serotonin levels during the observed 24 h interval. Moreover, all median values were within the reference range for serum serotonin (90.7 ng/mL, 108.0 ng/mL, 105.6 ng/mL, 87.6 ng/mL). Compared to serum serotonin levels, the currently used biomarkers for myocardial damage, myoglobin, troponin, and CK, showed a significant increase after myocardial infarction and PCI (Figure 1). Appl. Sci. 2020, 10, x FOR PEER REVIEW 6 of 13 , troponin (C), and CK (D) were analyzed. After STEMI, serum serotonin showed no increase or decrease within the 24 h interval compared to the control group or compared to t = 0 (A). Myoglobin, troponin, and CK showed a significant increase after coronary angiography compared to t = 0 and compared to the control group. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to t = 0, * p < 0.05, ** p < 0.01, *** p < 0.001 compared to control group.

Analysis of Thrombolysis in Myocardial Infarction (TIMI) Grade
We analyzed the thrombolysis in myocardial infarction (TIMI) grade prior to and post coronary angiography. Whereas most patients presented a TIMI grade of 0 or 1 prior to coronary intervention, it was 2 to 3 in all of them post PCI, indicating successful recanalization of the diseased vessel. Serum serotonin levels did not differ between the TIMI grades ( Figure 2).  , troponin (C), and CK (D) were analyzed. After STEMI, serum serotonin showed no increase or decrease within the 24 h interval compared to the control group or compared to t = 0 (A). Myoglobin, troponin, and CK showed a significant increase after coronary angiography compared to t = 0 and compared to the control group. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to t = 0, * p < 0.05, ** p < 0.01, *** p < 0.001 compared to control group.

Analysis of Thrombolysis in Myocardial Infarction (TIMI) Grade
We analyzed the thrombolysis in myocardial infarction (TIMI) grade prior to and post coronary angiography. Whereas most patients presented a TIMI grade of 0 or 1 prior to coronary intervention, it was 2 to 3 in all of them post PCI, indicating successful recanalization of the diseased vessel. Serum serotonin levels did not differ between the TIMI grades ( Figure 2). , troponin (C), and CK (D) were analyzed. After STEMI, serum serotonin showed no increase or decrease within the 24 h interval compared to the control group or compared to t = 0 (A). Myoglobin, troponin, and CK showed a significant increase after coronary angiography compared to t = 0 and compared to the control group. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to t = 0, * p < 0.05, ** p < 0.01, *** p < 0.001 compared to control group.

Analysis of Thrombolysis in Myocardial Infarction (TIMI) Grade
We analyzed the thrombolysis in myocardial infarction (TIMI) grade prior to and post coronary angiography. Whereas most patients presented a TIMI grade of 0 or 1 prior to coronary intervention, it was 2 to 3 in all of them post PCI, indicating successful recanalization of the diseased vessel. Serum serotonin levels did not differ between the TIMI grades ( Figure 2).

Ischemia/Reperfusion Injury
Patients who were deemed to have suffered from I/R injury after PCI regarding the ECG analysis displayed no difference in mean serum serotonin levels in comparison to those deemed to have achieved adequate reperfusion (109.1 ± 84.08 pg/mL vs. 97.52 ± 58.8; p = 0.9663), Figure 3A.

Ischemia/Reperfusion Injury
Patients who were deemed to have suffered from I/R injury after PCI regarding the ECG analysis displayed no difference in mean serum serotonin levels in comparison to those deemed to have achieved adequate reperfusion (109.1 ± 84.08 pg/mL vs. 97.52 ± 58.8; p = 0.9663), Figure 3A.
The analysis of the TIMI myocardial perfusion grade (TMP) and the myocardial blush grade (MGB) did not show altered serum serotonin levels in patients with I/R injury if assessed visually with these markers (Figure 2B-C).    The analysis of the TIMI myocardial perfusion grade (TMP) and the myocardial blush grade (MGB) did not show altered serum serotonin levels in patients with I/R injury if assessed visually with these markers (Figure 2B,C). Figure 4 demonstrates that the severity of disease and the outcome were not associated with serum serotonin levels. Neither the type of coronary artery disease (one-, two-, or three-vessel disease) nor the left ventricular ejection fraction (LVEF) were associated with different serum serotonin concentrations.

I/R injury no I/R injury
Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 of 13

Ischemia/Reperfusion Injury
Patients who were deemed to have suffered from I/R injury after PCI regarding the ECG analysis displayed no difference in mean serum serotonin levels in comparison to those deemed to have achieved adequate reperfusion (109.1 ± 84.08 pg/mL vs. 97.52 ± 58.8; p = 0.9663), Figure 3A.
The analysis of the TIMI myocardial perfusion grade (TMP) and the myocardial blush grade (MGB) did not show altered serum serotonin levels in patients with I/R injury if assessed visually with these markers (Figure 2B-C).

Correlation Analysis
After Bonferroni's correction for multiple comparisons (new significant p-value set at 0.0022, Table 3), none of the characteristics listed demonstrated a significant correlation with serum serotonin levels.

Discussion
In this pilot study, we investigated for the first time if 5-HT serum levels are altered in patients with STEMI and if they are associated with the severity of myocardial damage, in order to evaluate it as a potentially novel biomarker for AMI and I/R injury.
Immediate restoration of blood flow by coronary angiography improves survival and outcome in patients with STEMI [3,4]. The death of cardiomyocytes during STEMI results from both necrosis and apoptosis. Troponins and CK-MB are routinely used in clinical practice and are validated biomarkers of cell lysis, necrosis, and overall myocardial damage to estimate infarct size [6,35,36]. However, the specificity of these markers is limited [7] and they can be elevated in acute myocardial injury of non-ischemic origin [6,8,9]. Moreover, troponins can be chronically raised in patients with congestive heart failure or chronic kidney disease [6,7] and CK and myoglobin can be altered in patients with rhabdomyolysis [37]. Therefore, there is need for additional, more specific, and more precise biomarkers.
In this pilot study, we assessed if serum 5-HT correlates with clinical variables and markers of cell lysis (e.g., CK, myoglobin, and cTNT) and further investigated whether it could predict I/R injury in STEMI patients after PCI. The hypothesis that 5-HT serum levels might be associated with the extent of myocardial damage after AMI was, on the one hand, based on previous studies reporting altered serotonin levels after myocardial infarction and, on the other hand, based on recent findings that that 5-HT aggravates I/R injury [12,22]. 5-HT plays a crucial role in AMI. Ninety to ninety-five percent of 5-HT is stored in the dense granules of platelets, and released upon activation [13]. In AMI, the platelet-stimulating microenvironment and the thrombus formation in the coronaries lead to an activation of platelets with the subsequent local release of 5-HT [22]. The increased local 5-HT levels contribute, on the one hand, to enhanced recruitment and aggregation of platelets [13]. On the other hand, 5-HT, as a vasoactive substance, leads to the constriction of coronary smooth muscle cells, further worsening myocardial damage [23]. Consistent with this are findings from rat experiments revealing an increase in 5-HT concentrations in the heart tissue after myocardial infarction [23], representing enhanced platelet activation and aggregation. As a consequence of the release of platelet 5-HT in the coronary arteries, plasma 5-HT levels markedly increase after AMI in mice and humans [12,25]. Mauler et al. reported a significant increase in plasma 5-HT in mice within the first 24 h after LAD ligation [12]. In humans, this increase in plasma 5-HT could also be observed: Jin et al. studied 30 cases of AMI and compared them with 13 cases of unstable angina. They reported a rapid increase in plasma 5-HT within the first 24 h after AMI, compared to no alteration in the unstable angina group. Moreover, they found an association between the severity of AMI and plasma 5-HT levels [25]. As discussed by Brand and Anderson [38] and as we experienced in previous experiments, the accurate generation of plasma for the determination of 5-HT levels is delicate and easily prone to error. Plasma concentrations of 5-HT are very low under physiologic conditions [13] and any activation of platelets during the process of plasma generation needs to be strictly avoided. To ensure that, immediate and gentle processing of blood samples is necessary, which is not always possible in clinical practice. We therefore aimed to evaluate serum 5-HT as a potential biomarker, which is much easier to handle and less prone to error. However, we could neither find differences in serum serotonin levels between our STEMI cohort and the control group nor find any changes in 5-HT levels detectable within the 24 h study period in our collective.
A second objective of our pilot study was to examine whether serum 5-HT was suitable as a biomarker for I/R injury. As mentioned above, early reperfusion is mandatory in patients with STEMI to salvage myocardium and reduce infarct size. However, reperfusion itself might provoke additional injury, leading to increased infarct size and microvascular dysfunction [10,39]. This deleterious process of reperfusion is characterized by ionic imbalance [40][41][42] and excess formation of reactive oxygen species [41], both contributing to necrosis, apoptosis, and cell death [10]. All these processes lead to impaired myocardial blood flow, resulting in the no-reflow phenomenon, myocardial stunning, and reperfusion arrhythmias [10]. About 35% of STEMI patients suffer from incomplete reperfusion due to severe I/R injury [43]. Currently, myocardial damage induced by I/R injury cannot be monitored due to a current lack of specific biomarkers.
It is well established that 5-HT causes deleterious effects in I/R injury: In 1994, it was first demonstrated that administration of nexopamil, a combined Ca 2+ and 5-HT antagonist, reduced myocardial damage in a pig model after LAD occlusion [44]. In 2014, Guner et al. showed that the administration of fluoxetine, a selective 5-HT reuptake inhibitor, prevented I/R injury after aortic surgery in rats [45]. It was claimed that 5-HT deteriorates I/R injury by enhancing oxidative stress caused by mitochondrial monoaminooxidase-A (MAO-A) activity during the process of 5-HT degradation [46]. Mauler et al. recently found that 5-HT aggravates myocardial I/R injury by aggravating neutrophil degranulation, and that I/R injury was accompanied by an elevation in 5-HT plasma levels [12]. Moreover, when Tph1 −/− mice (mice without peripheral serotonin) were subjected to myocardial ischemia with reperfusion, the infarct size was significantly reduced by about 35% compared to wildtype (WT) controls [12]. We therefore hypothesized that serum 5-HT levels might be associated with I/R injury in our collective.
We used the comparison of ST-segment elevation in standard 12-lead ECG prior to and after PCI as a surrogate marker for I/R injury. Incomplete resolution of ST-segment elevation has been shown to be an indicator for I/R injury and a prognostic marker for 90-day prognosis, death, development of heart failure, and shock [30,31]. Moreover, we additionally analyzed the visual markers TMP and MGB for the assessment of I/R injury: Impaired perfusion of the myocardium in coronary arteriography assessed by use of the TMP grade was described to be related to a higher risk of mortality [47]. Moreover, the outcome assessed by the MBG was also associated with mortality in patients with STEMI [34]. Contrary to our expectations, 5-HT serum levels did not correlate with I/R injury in our collective. We assume that the local release of 5-HT in the coronary vessels is too low to alter systemic serum 5-HT concentrations.
Our pilot study was limited by the relatively small sample size of patients with STEMI. It was clearly underpowered to identify small differences in serum 5-HT and its association with I/R injury. However, CK and cTnT, as established biomarkers for AMI, increased as expected, whereas serum 5-HT did not show any kinetics, indicating a lack of suitability of 5-HT as an appropriate biomarker.
Another strength of the study is that blood was drawn at four distinct time points after myocardial infarction. This allowed for a time course of 5-HT serum levels during AMI and again revealed that it is not suitable as a biomarker for the extent of myocardial damage or I/R injury.

Conclusions
In conclusion, we have shown that 5-HT serum levels are not suitable as a biomarker in acute myocardial infarction. There was no alteration in serum serotonin levels after STEMI compared to an age-matched control cohort. We did not observe any timeline change within 24 h post PCI compared to before PCI. Moreover, the severity of coronary artery disease, the outcome of coronary angiography, the extent of ischemia/reperfusion injury, and the degree of heart failure were not associated with any alterations in 5-HT serum levels.