ABSTRACT
Aim:
Carbon monoxide (CO) poisoning is still an important factor in the rate of emergency visits. For this reason, we aimed to compare the incidence of myocardial infarction (MI) late after discharge of CO poisoning and its types as compared with the normal population.
Materials and Methods:
A total of 1369 patients with a diagnosis of CO intoxication and 1617 patients without a history of cardiac disease who were admitted to the emergency department between January 2005 and December 2010 were included in the study. Patients with a COHb level above 10% and the control group was followed up for 60 months for MI. The patients were divided into three groups: inferior, anterior, and non-ST elevation MI.
Results:
At the end of sixty months of follow-up, MI was determined in 103 (7.52%) of the CO group and 61 (3.77%) of the control group. When both groups were considered together, a significant relationship was found with gender, mortality, diabetes, hypertension, and use of tobacco products. However, when the CO group was evaluated sequentially, it was seen that it was only associated with gender, mortality, and MI types. MI groups were statistically significant with other variables except gender. In univariate and multivariate linear regression analyses, age, COHb, Tn level, and CO exposure time was found to cause increased mortality and risk of MI. After follow-up, inferior MI and mortality were higher in the patient group, and anterior MI and mortality were higher in the control group.
Conclusion:
Acute COHb and Tn levels may be important values in defining the risk of late MI development in patients discharged after CO poisoning.
Introduction
Carbon monoxide (CO) poisoning, which is the leading cause of toxic deaths, constitutes an importImportantof emergency department (ED) admissions (1,2). Carboxyhemoglobin (COHb), which is formed by binding to hemoglobin with an affinity 240 times higher than oxygen, disrupts the distribution of oxygen to textures and carrying of oxygen Allosteric change occurs when uni of the CO binds to the heme part of the hemoglobin. The binding of the residual three oxygens to Heme increases and the emancipation of oxygen to tissues decreases (3,4). CO has more affinity for cardiac myoglobin than hemoglobin. Therefore, myocardial depression and hypotension may occur due to tissue hypoxia. CO combines with myoglobin and causes a decrease in partial oxygen in muscle tissue and ultimately causes rhabdomyolysis (4-6). Myocardial infarction (MI) can often be associated with CO exposure.
Even 5-10% increases in COHb level in people with coronary disease before can trigger Angina that occurs during exercise. High levels of COHb can lead to myocardial depression even in young and healthy individuals (6). Electrocardiogram (ECG) and cardiac troponin (Tn) should be studied in order not to overlook silent ischemia. Tns are sensitive and specific markers of heart muscle damage. In 2000, they were accepted as standard markers in the diagnosis of acute MI (7-10).
In the study, it was aimed to evaluate the relationship between Tn and COHb levels with the types of MI who presented with ED because of CO poisoning and that could develop in the late period after the patient was discharged.
Materials and Methods
In this cross-sectional cohort study, two groups above 18 years old who were referred to the ED between January 2005 and December 2010 were included. While the patients diagnosed with CO poisoning formed the patient group, and those who had recently been admitted to the emergency service with CO poisoning and had no history of cardiac disease formed the control group. While determining the control group, attention was paid to the age of the patients and to be close to the age of the cases diagnosed with CO poisoning.
Patients diagnosed with CO poisoning were divided into two groups at late-period (1-60 months) as acute coronary syndrome (ACS) development and non-development after discharge. Patients with ACS were divided into three groups inferior MI group (inferior, right ventricular, inferolateral, inferoposterior and posterior MI), anterior MI group (septal, anterior, lateral, high lateral and diffuse anterior MI), and non-ST elevated MI group.
Patients with previous diagnosis of infectious or inflammatory disease or malignancy, previous treatment for severe anemia or other hematological diseases or anemia, and those given erythrocyte suspension in the last six months were excluded from the search. After the patients were discharged from CO poisoning, they were followed up retrospectively for 60 months with an annual automation system.
Statistical Analysis
Data obtained in the study were analyzed using IBM Statistical Package for the Social Sciences Statistics for Windows 20.0 (IBM Corp., Armonk, NY, USA). The Shapiro-Wilk test was used to analyze the normal distribution of the variables. The Student’s t-test was used for the variables with normal distribution, and the Mann-Whitney U test was used when examining the differences between the groups for those with non-normal distribution. Chi-square analysis was performed to examine the relationships between the nominal variable groups. Correlation analysis was performed using CO poisoning patient and control groups variables. In addition, univariate linear regression analysis was performed with all the variables in the CO poisoning patient and control groups variables. Predictive values were determined by multivariate linear regression analysis for the significant parameters in the univariate analysis. When interpreting the results, p<0.05 was considered statistically significant.
Results
The mean age of the patients was 51.71±17.62 years (52% male, 45.8% patients), and the mean follow-up period was 60 months (range 19 to 88). After sixty months of follow-up, MI was observed in 103 (7.52%) of 1369 patients in the CO group and in 61 (3.7%) of 1617 patients in the control group. The clinical and demographic characteristics of the patients are noted in Table 1.
After sixty months of follow-up, CO poisoning and control groups were found to be statistically significant with developing ACS and diabetes mellitus, hypertension, tobacco use, gender, mortality, and MI types (p=0.001, Table 2).
In the chi-square analysis between ACS and variables that developed after CO poisoning, diabetes mellitus (p=0.091), hypertension (p=0.954), and tobacco use (p=0.621). Were found to be statistically significant with gender, mortality, and MI types (p=0.001, Table 3).
As a result of sixty months of follow-up, in the chi-square analysis of MI types with variables, it was found to be statistically significant with gender (p=0.061), mortality, control and patient groups, hypertension, diabetes mellitus, and use of tobacco products (p=0.001, Table 4).
In univariate analysis, age, emergency admission time, mortality, aspartate amino transferase, alanine aminotransferase (ALT), alkaline phosphatase, C-reactive protein (CRP), Tn, and COHb levels were found to be prognostic indicators. However, in multivariate linear regression analysis, factors including age, the duration of emergency admission, mortality, ALT, CRP, Tn, and COHb were associated with increased MI risk (Table 5).
The correlation between Tn, COHb, mortality, exposure time to CO, and MI types with 60-month cardiac follow-up after CO poisoning was statistically significant (p=0.001, Table 6).
Discussion
In this study, we tried to determine MI types and prognosis after CO poisoning. CO poisoning can be acute or chronic. Since CO has a higher affinity to hemoglobin than oxygen, oxygen is delivered to tissues in fewer amounts. Thus cardiac toxicity may cause myocardial hypoxia. However, the direct toxic effect on myocardial mitochondria plays a more important role (11-14).
In the case of CO poisoning, the cause of cardiac damage is based on 2 mechanisms. The first is ischemic devastation provoked by COHb binding to heme proteins in lieu of oxygen, and the second is toxic damage directly induced by CO (2,15-19). While the cell undergoes direct toxic damage at the mitochondrial level with CO, cytochrome c oxidase inhibition and decreased glutathione levels occur (20,21). This results in anaerobic metabolism in cardiac myocytes, resulting in hypoxia, lactic acidosis, and apoptosis. The cause of endothelial damage is the induction of the enzyme that occurs during apoptosis formation (22). In contrast, CO stimulates the oxidation of low-density lipoproteins and this increases the free radical formation by inducing peroxynitrite nascency in plasma (23,24). It has been shown that CO exposure triggers venous, arterial, and even stent thrombosis and has a prothrombotic effect (25-30).
In many studies conducted so far, gender and age vary. There were 1441 (48.2%) men in our study and the average age were 51.72 years. In the control group, there were 630 (38.9%) males and their mean age was 63.33 years, and in the patient group, 810 (59%) males and their mean age was 36.2 years. Türkmen and Akgöz (31) found 61.62% males and an average age of 37.73 years in their study; in the Durak (32) study, 71.83% males and 33.39 years of average age were determined. However, Hosseininejad et al. (33) in the meta-analysis study found that of the 4620 people included in the study, 40.12% were male and the average age was 31.68 years. The average age in our study was slightly higher. This may be due to the family and social structure and the other to the general increase in the age average of society. Cardiac involvement can occur immediately after exposure to CO or a few days later. Arrhythmias such as palpitations, sinus tachycardia, atrial fibrillation, and ventricular extrasystole may be observed. In severe cases, bradycardia and complete atrioventricular block may be seen (34). In those with ischemic heart disease, angina pectoris and MI can be triggered. ECG ST segment and T wave changes are common. Transient right and/or left ventricular wall motion disorders may be present (35). In the study published by Lippi et al. (36), the damage caused by CO to the cell; it has been reported that it may cause various cardiac clinical pictures such as cardiomyopathy, angina, MI, arrhythmia, heart failure, pulmonary edema, cardiogenic shock and sudden death. Cases of silent MI caused by acute CO poisoning without chest pain have been reported in the literature. CO cardiotoxicity may be clinically latent and often remain undiagnosed due to specific ischemic changes in ECG and inadequate symptoms (37,38).
There are many studies on ACSs. However, there is no clear information about MI frequency in studies. Sanchis-Gomar et al. (39) found the frequency of MI to be 1.1% in women and 1.7% in men. However, Turkey Cardiovascular Diseases Prevention and Control Program Action Plan (40) women in the study 1.33%, 3.03% in men was found to be acute MI. In our study, patients discharged after CO poisoning was followed for 60 months. These data were retrospectively recorded annually. After CO poisoning, MI types that were not previously detected in the literature were examined. At the end of this 60-month follow-up, it was observed that MI developed in 103 (7.52%) patients in the CO group. In spite of the higher mean age of the control group 61 (3.7%) cases had MI. The 2-fold higher MI frequency may be a predictive indicator of MI that may develop CO intoxication. Tn and COHb values in all patients of the Co patient group were significantly higher than those without MI at the 60-month follow-up. Inferior MI in the CO patient group and anterior MI in the control group was more common. While most MI cases in the control group were diagnosed with hypertension, most cases developing MI after intoxication was normotensive patients. This could be due to two reasons. First, the inferior group MI is generally hypotensive and bradycardic. Second, the sinus node is close to the right coronary artery. It was most common in mortality in MI with ST elevation, especially in the inferior MI group.
Henry et al. (41) found that acute myocardial damage in patients with CO poisoning, followed for an average of 7.6 years, is a long-term risk of death. In another study with 10.6 years follow-up after CO poisoning, it was found to be a neutral predictor of all long-term mortal situations. Huang et al. (42) found that CO intoxication increases the risk of death in the long term. Lee et al. (43) showed that CO intoxication increases the risk of CAD and heart failure in the long term.
In our study, during the 60-month follow-up, mortality was detected as 31 (2.2%) in the CO patient group and 19 (1.1%) in the control group. In all cases with mortality in the CO patient group, it was determined that the COHb level was above 40 and the accompanying Tn elevation. Mortality in the CO patient group was common in the inferior MI group, and in the control group, anterior MI was common. The correlation between MI and mortality was positively correlated.
This is the first study to establish whether high co levels are a long-term detached risk factor for MI types. Kalay et al. (44) emphasized that increased CO levels in blood and extended duration of exposure to CO poisoning are risk factors for MI development. In our study, the exposure time and rate of cardiac damage were similar.
As a result, the high COHb and Tn values in the acute period of the patients who applied to the ED with CO poisoning and the MI frequency in the late period after discharge were more than two times higher than the normal population. In addition, while anterior MI and mortality were common in the normal population, inferior MI and mortality were found to be higher in the CO patient group. These data show that patients are in a higher risk group after CO discharge.
Study Limitations
The major limitation of the study is that it is not multi-centered and retrospective. Therefore, it was difficult to reach some data.
Conclusion
Cardiac disorders that may develop in the future can be easily overlooked in CO poisoning, especially if the patients are asymptomatic. Considering the COHb and Tn levels evaluated in our study by physicians may guide the prevention of future cardiac problems and early diagnosis.
Ethics
Ethics Committee Approval: The study was approved by the Cumhuriyet University Faculty of Medicine of Local Ethics Committee (decision no: 2012/12-09, date: 04.12.2012).
Informed Consent: Retrospective study.
Peer-review: Externally and internally peer-reviewed.
Financial Disclosure: The author declared that this study received no financial support.