A Retrospective Study Assessing the Effect of Infections in Myocardial Infarction Patients: Mortality and Complications

Background: There is a well-established relationship between myocardial infarction and infection. Multiple articles describe the increased risk of myocardial infarction, both type 1 and 2, following an infectious process. However few articles have described the relation between concomitant myocardial infarction and infections on same admission mortality and complications. Methods: The aim is to assess the effect of an acquired or concomitant infection on complications and mortality during hospitalized cases of myocardial infarction. 1197 patients of different types of myocardial infarction were studied in correlation to infectious processes. Cultures from different sites were collected and isolation of various bacterial agents were studied. Mortality and various complications were compared between infected and non-infected subjects. Pearson's chi squared test was used to compare percentages (or the Fisher exact test when expected values were lower than 5). Moreover, means were compared through ANOVA, after checking data normality and homoscedasticity. A likelihood ratio backward stepwise method was used to conduct dichotomous logistic regressions, taking dichotomous outcomes as dependent variables, and sociodemographic and biological characteristics as independent variables (potential confounders). Results: Wound, sputum, blood and urine infections were associated with increased same admission mortality and complications. Microorganisms were then studied alone regardless of the site of infection and it was shown that Escherichia Coli, Escherichia Coli ESBL, Candida Albicans, Pseudomonas Aeruginosa and Staphylococcus of any type were signicantly associated with same admission complications when associated with myocardial infarction. Length of stay was signicantly elevated in patients with concomitant infection and it increased with the addition of positive cultures from different sites. Conclusion: Concomitant infections with myocardial infarction signicantly increase the risk of same admission complications, mortality and length of stay regardless of the site of infection and type of microorganisms. longest hospitals stays were found in patients with positive blood cultures. Patients with myocardial infarction had longer hospital stays when they had positive cultures at multiple sites and the longest stays were found in patients with three positive cultures at different sites. Patients infected by Acinetobacter Baumanii had the longest duration of stay followed by Escherichia Coli ESBL.

and reviews discuss post-infectious myocardial infarctions. However, studies must be conducted to assess the effect of infections on the course of disease in myocardial infarction patients.

Objective
The objective of this study is to compare mortality and complication rates in myocardial infarction (MI) patients with and without concurrent infections. We hypothesized infections in association with myocardial infarction contribute to higher mortality rates as compared to myocardial infarction alone regardless if the infections is associated with symptoms or is asymptomatic. We also aimed to study each microorganism regardless of the infection site and try to link the causative infectious agent with the pattern of increased mortality and complications.

Study Design:
This work is a retrospective study. Data was collected from hospitalized patients from 2009 to 2020 in a single center, Mount Lebanon Hospital University Medical Center. Patients were randomly selected based on the nal diagnosis on the discharge paper.

Participants:
Patients included were diagnosed of having acute myocardial infarction of any type.
Myocardial infarction was de ned as ST elevation myocardial infarction, non-ST elevation myocardial infarction and unstable angina.

Data Measurement:
Basic information was collected about patient's characteristics, duration of stay in the hospital, admission unit. Patients were also evaluated based on the presenting symptoms and signs. Data concerning the given treatment was also collected including medical therapy or emergent cardiac catheterization interventions. Data on mortality and complications was noted in addition to basic laboratory tests such as troponin, CRP, creatinine and platelets level. Different types of cultures were recorded, and data from wound, sputum, blood, and urine culture were collected in order to nd a correlation with mortality and complications. The duration of hospital stay was evaluated in regard of positive cultures and presence or absence of a speci c microorganism.
This trial was approved by Institutional Review Board IRB from the hospital research center.

Statistical Methods:
Data was analyzed using SPSS software, version 25.0. A descriptive part helped to summarize multiple data variables. Minimal and maximal values were collected with mean and standard deviations analysis. Pearson's chi squared test was used to compare percentages (or the Fisher exact test when expected values were lower than 5). Moreover, means were compared through ANOVA, after checking data normality and homoscedasticity. A likelihood ratio backward stepwise method was used to conduct dichotomous logistic regressions, taking dichotomous outcomes as dependent variables, and sociodemographic and biological characteristics as independent variables (potential confounders). Adjusted OR were calculated, along with their 95% con dence interval.
The baseline characteristics of patients concerning age, duration of stay, admission systolic blood pressure, diastolic blood pressure, temperature, troponin level and CRP level are summarized in Table 1. Social history of patients was evaluated and 56.47% of patients were active smokers, 16.21% were excessive caffeine consumers and 7.69% were active alcoholics ( Table 2).   All causes complications were also studied in comparison to concomitant urine culture infections. From the 121 patients that developed urine infection, 67 had same admission complications. The association was signi cant with a p value of 0.001 (Table 4.a).
The most common cause of death was cardiogenic shock in 124 cases followed by septic shock in 34 cases. Ventricular arrhythmias occurred in 6 patients while pericardial tamponade and complete AV block occurred in 2 and 1 patients respectively. Most patients had combination of causes, especially patients with concomitant infections with high risk of septic shock complicating cardiogenic shock and leading to higher mortality rates.
Microorganisms were then subdivided by type of bacteria, virus, or fungi and regardless of site of infection.
The presence of a speci c microorganism was rst studied in relation to myocardial infarction type and the percentages of each microorganism was calculated in all groups of myocardial infarctions. E. coli was the most prevalent bacteria in patients with NSTEMI and unstable angina (p=0.041) while Candida albicans was most prevalent in case of STEMI (Table 5.a). The same microorganisms were studied in relation to all causes mortality and all causes complications during the same admission (Table 5.b). The presence of Pseudomonas, Escherichia coli ESBL, Candida albicans, Escherichia coli staphylococcus and streptococcus at any site were associated with same admission complications of any type and the correlation was signi cant.
A signi cant correlation was established between death and the presence of pseudomonas, Escherichia coli ESBL, Candida albicans, klebsiella pneumonia, E. coli, Enterobacter, Acinetobacter, and Staphylococcus at any site (Table 5.b).  The effect of infections on hospital stay was studied and it was found that patients with myocardial infarction of any type had an average of 7.5 days (SD = 8.9 days) before outcome (discharge home or death) ( Table 1). When excluding patients that developed infections during the same admission, the average hospital stay for patients with myocardial infarction without evidence of infections dropped to 6.0 days (SD = 5.

Multivariable analysis
When studying the relation between mortality in patients with infection with other factors (Table 7a), it was shown that all these factors were signi cantly associated with increased mortality, except for high and very high blood pressure (high blood pressure was considered in our study as a systolic blood pressure between 140 mmHg and 160 mmHg and very high blood pressure as systolic blood pressure above 170 mmHg). Also, alcohol consumption gave a borderline result. Complications post MI were signi cantly associated with the factors reported in the model, except for an inverse association with High and very high blood pressure, high uric acid (all levels above 7 were considered as a high uric acid in our study), and chest pain, that were associated with lower risk of complications (Table 7b).

Discussion
Acute coronary occlusion will lead to myocardial infarction referred as type 1 MI [11]. Infections are postulated to be a trigger factor for the rapid development of atherosclerosis [12]. Autopsies done in animal models have shown that infections and systemic in ammations have caused increased atheromatous activities in coronary plaques [13]. Various cytokines and in ammatory mediators play a role such metalloproteinase and peptidases and they activate oxidative burst which will lead to unstable plaques [14].
Infections induce a hypercoagulable state promoting coronary thrombosis mainly in patients with underlying coronary atherosclerosis [15]. This mechanism is promoted by activation of neutrophils and platelets, in addition to activation of tissue factors, impaired coagulation cascade and alteration of endothelium function [16]. It was proven that patients with in uenza and pneumonia had increased in myocardial infarction rates due to activation of the hypercoagulable mediators [17]. This state of in ammation and thrombosis persisted beyond the infection duration and the risk of myocardial infarction persisted for longer periods [18].
Type 2 myocardial infarction occurs in case of increase in oxygen demands [11]. Infections and sepsis will cause lower blood pressure and increase in heart rates which will shorten diastole [19]. Coronary perfusion occurs in diastole and in case of infection, the coronary perfusion will be lower [20]. In cases of pneumonia, lower oxygen levels will impact myocardial oxygenation and promote ischemia leading to type 2 myocardial infarction [21].
Autopsies in animal models showed that ischemia in infections may occur due to increased myocyte vacuolization leading to loss of myocytes rather than coronary artery stenosis increasing the risk of arrhythmias and heart failure [14].
Patients who had received vaccines for different infections were shown to have decrease in myocardial infarction risk due to decrease in the in ammatory mediators and hypercoagulable state [22]. Patients were started on appropriate antithrombotic therapy and 467 received antibiotics due to simultaneous infection even if absence of positive cultures. The mortality rate during the same admission was 13.0%, 156 patients had death outcome and the most common cause of death was cardiogenic shock in 125 patients followed by septic shock in 35 patients. Two patients died from complete heart block, two from ventricular arrythmias, two from pericardial tamponade and the cause of death was unspeci ed in two cases. The causes of mortality were similar to mortality in isolated myocardial infarction without the presence of infections with cardiogenic shock being the established most common cause of death during the same admission.
Complications without leading to mortality during the same admission were also studied and the most common being acute kidney injury seen followed by cardiogenic shock, pneumonia, DIC, pulmonary edema, atrial brillation, ventricular arrhythmias, and other less common complications. All complications were grouped in the same category of all causes complications, and it was con rmed that positive infections and cultures in wound, blood, sputum and urine were associated with increased risk of complications during the same admission with a signi cant p value. Acute kidney injury was very common, because infections and sepsis predispose the patient to hypotension and systemic vasodilation causing decrease renal perfusion and activation of the renin angiotensin aldosterone system [23]. This condition predisposes to prerenal azotemia and if remained untreated will lead to intrinsic renal damage and acute tubular necrosis leading to acute renal failure and many patients were started on dialysis consequently [24]. Cardiogenic shock was also signi cantly associated with patients that developed simultaneous infection with myocardial infarction. As in the case of renal failure, systemic infections will lead to systemic vasodilation in cases of septic shock leading to compensatory tachycardia as a response to preserve cardiac output [25]. The myocardium will compensate early in the course of the disease but due to the additional strain from coronary artery disease, the heart muscles will fail and lead to cardiogenic shock requiring vasopressors and inotropes [26]. Cardiogenic shock will lead to pulmonary edema and ventricular arrhythmias exacerbation the condition [27].
Several studies were published regarding complications and mortality in myocardial infarction with concomitant infections and revealed results relatively in line with ours. A recent large study including 10,880,856 AMI admissions among which 745,536 cases of concomitant respiratory infections done by Vallabhajosyula et al. (2021) concluded that respiratory infections signi cantly impact AMI admissions with higher rates of complications, mortality, and resource utilization [28].
Concomitant in uenza respiratory infection increased risk mortality and multi-organ failure in MI patients as well, however it was associated with decreased rate of cardiac angiography and revascularization events which was not investigated in our study (Cardoso et al., 2020) [29].
Lower utilization of percutaneous coronary intervention and coronary angiography was observed also by Parenica et al. (2017) in their observational study on acute MI patients presenting with cardiogenic shock as well [30]. But in contrast to the earlier literature ndings and the ones in our study, Parenica et al. (2017) found no signi cant difference in mortality between infected and non-infected acute MI patients admitted with cardiogenic shock [30]. In fact, Patlolla et al. (2021) in their recent published article observed lower in-hospital mortality rate in acute myocardial infarction and cardiogenic shock patient complicated by respiratory infection compared to those with no infections, as well as lower utilization of coronary angiography and PCI which was attributed to higher rates NSTEMI-CS and lower rates of STEMI-CS in admissions with respiratory infections compared to those without [31].
Worse prognosis was not only limited to respiratory infections in the literature. While in our study, blood positive culture was relatively less common compared to sputum cultures, in other observational study of serious infections following myocardial infection, most of the infections were found to be isolated from the blood and demonstrated worse clinical outcome and longer hospital stay, which indicate that worse outcome is related to the infections in general and not only isolated to respiratory ones [9].
The second objective was to subdivide the microorganism by type regardless of the site of infection and correlate these ndings with mortality and complications during the same admission. The reason of this selection is to try to identify a common link between the same microorganism present at different site and to detect a correlation with mortality. Some microorganisms secrete substances or toxins that can diffuse to the systemic circulation regardless of the site of origin. These substances will have an impact on the cardiac function leading to the development of complications and death. From the microorganisms studied in Table 6, only Enterococcus, Enterobacter and Acinetobacter were not signi cantly associated with the development of complications during the same admission, although Acinetobacter infections were strongly linked with mortality mainly in patients on mechanical ventilations and in the intensive care unit. Only streptococcus and enterococcus were not signi cantly correlated with mortality during the same admission with a p value of more than 0.05. To the best of our knowledge, no previous studies were done to investigate the effect of speci c organisms in myocardial infarction patient who developed infectious process during their hospital stay.
Hospital length of stay is an important factor regarding cost of hospitalization, complications, and mortality. It is well established that the longer hospital stay is associated with major complications regardless of the initial diagnosis and the aim is to reduce the length of stay to promote and improve the health care system. Patients with myocardial infarction but with no evidence of infections or positive cultures were compared to patients with positive cultures in a single site, two sites and three sites. To note that when studying the length of stay, A total of 100 patients developed sputum infection, and the duration of stay for these patients were on average 15.3 days, which is more than double the duration spent by MI patients without any infections (6 days). These results were in line with those found in previous studies demonstrating that indeed MI admissions with concomitant respiratory infection presenting with in uenza [29] or with cardiogenic shock [31] have longer hospital stay and higher cost compared to those with no respiratory infections. Duration of hospital stay for patient with concomitant respiratory infection was slightly lower than those with bloodstream infections but higher than those with wound infections (45 patients averaging 12.0 days) and urine infections (113 patients averaging 11.7 days). Wound infections and urine infections are less severe types of infections that require lower antibiotics therapy duration and are a less frequent cause of septic shock as compared to blood and sputum culture infections.
Three germs, common in the intensive care setting, when studied regardless the positive culture site, showed that Acinetobacter baumanii was associated with the longest hospital stay reaching 25 days (total number of patients with this infection and myocardial infarction is 10 and all patients were admitted to the intensive care unit). Such infection requires broad spectrum antibiotics and are often resistant to most antimicrobial agents, requiring sometimes switching between therapies and pharmacological stewardship to decrease the bacterial resistance. These measures are essential for patient safety and to decrease the length of stay.
When using multivariable analysis, the mortality in patients with myocardial infarction was affected by many factors. But it was shown that patients with high or very high blood pressures (>140 mmHg) had lower level of mortality. This is potentially due to the fact that patients with higher blood pressures in the onset of myocardial infarction are treated more aggressively. By lowering blood pressure, afterload is lowered and myocardial strain will improve thus leading to improve outcomes. Similarly complications were associated with multiple factors but it was shown that higher blood pressure was a protective variable, similar to high uric acids. Patients with chest pain were found to have lower complications than patient without pain. Silent myocardial infarctions are more serious because of the lack of signs. In these cases, diagnosis is delayed or missed and the appropriate treatment is not administered, thus the higher incidence of complications in this subgroup.

Conclusion
Myocardial infarction is a serious and life-threatening condition that is associated with high rates of mortality.
Patients with this condition will always require hospital admissions and long stays. This condition can lead to serious complications such as arrhythmias and heart failure. When infections were simultaneously associated to myocardial infarction, it was proven that patients had increased in rates of same admission mortality and complications. The results were applicable for patients with wound, sputum, blood and urine cultures infections. Microorganisms were then studied alone regardless of the site of infection, and it was shown that Escherichia Coli, Escherichia Coli ESBL, Candida Albicans, Pseudomonas Aeruginosa and Staphylococcus of any type were signi cantly associated with same admission complications when associated with myocardial infarction. Escherichia Coli, Escherichia Coli ESBL, Candida Albicans, Pseudomonas Aeruginosa Staphylococcus of any type, Klebsiella Pneumonia, Enterobacter, and Acinetobacter Baumanii were associated with same admission mortality in patients with myocardial infarction.
Patients with myocardial infarction without evidence of associated infections had lower length of stays in hospitals as compared to patients with myocardial infarction associated with positive wound, sputum, blood and urine. The longest hospitals stays were found in patients with positive blood cultures. Patients with myocardial infarction had longer hospital stays when they had positive cultures at multiple sites and the longest stays were found in patients with three positive cultures at different sites. Patients infected by Acinetobacter Baumanii had the longest duration of stay followed by Escherichia Coli ESBL.
Infections when associated with myocardial infarction, signi cantly increase the same admission complication, mortality and length of stay. Patients with myocardial infarction should be closely followed with infection preventive measures in order to avoid the two diseases association.
Multivariable analysis showed that higher blood pressures, high uric acid and chest pain were associated with less mortality and complications in patients with concomitant infections

Limitations
This study included one small pool of patient from a single hospital. Larger population pools and further retrospective and prospective studies must be conducted to assure and con rm the results found in our study.
More research must aim to build a rm relationship between different microorganisms infections and there effect on the course myocardial infarction hospitalization.

Declarations
All methods were carried out in accordance with the Declaration of Helsinki.
Ethics approval and consent to participate: The approval of all the study protocol was given by the institutional research ethics committee at Mount Lebanon Hospital named "Ethics' Committee of Mount Lebanon Hospital" in October 5, 2020. The approval was given until the study ends. The Ethics Committee operates under the understanding of Good Clinical Practice (GCP) under the guidance of the Declaration of Helsinki. All participants signed an informed participation consent.
Patient informed consent: This retrospective study is based on recorded data owned by "Mount Lebanon Hospital" and approval to use the data was obtained by its Ethics committee "Ethics' Committee of Mount Lebanon Hospital".

Consent for publication
All authors have approved the manuscript and agree with submission to this Journal.