Effect of serum osmolality on 6-year survival rates in patients with acute myocardial infarction

Aim: In the present study, we aimed to evaluate the potential relationship between serum osmolality and mortality rates in a six year of follow-up in patients with a history of acute myocardial infarction. Material and Methods: A retrospective study was designed. Participants were the patients with a first attack AMI, who were referred to our tertiary referral center for angiography. The relationship between the biochemical values of patients who were hospitalized between the period January 2008 - June 2009 and their survival in six years was investigated. Clinical variables of baseline characteristics, in-hospital management, and in-hospital adverse outcomes were recorded. Results: Two hundred and four patients, 174 men (85%) and 30 women (15%), were included in the study. Median serum osmolality was 295.87 mOsm/kg. Mean follow-up time was 61.31±1.68 months. The best cut-off value of the plasma osmolality to predict the 6-year mortality was 303.94 mOsmol/kg. Conclusion: The higher the osmolality, the worse the six-year survival is in patients with first episode AMI even in the absence of diabetes mellitus and chronic kidney disease. We believe that hyperosmolality can be targeted in treatment and prevention efforts as well as its use when evaluating outcomes of the cardiac diseases.


Introduction
Cardiovascular diseases including coronary artery disease, acute myocardial infarction, and heart failure is the leading cause of mortality worldwide (1,2).
Increased serum osmolality is considered a risk for coronary artery disease (3,4). The presence and severity of coronary artery disease confirmed by angiography was found to be related with the serum osmolality (5).The negative effect of diabetes mellitus, renal insufficiency and both hyponatremia and hypernatremia on coronary arteries is well documented and they have long been known as a risk factor for coronary artery disease (6)(7)(8)(9).
Osmolality is a useful marker of hydration status and it can be calculated using fasting blood glucose, blood urea nitrogen (BUN) and sodium (Na) values (5,10,11).Studies suggested that in patients presenting with Acute Myocardial Infarction (AMI), hyperglycemia at admission is associated with increased mortality (12,13).Moreover, the relationship between impaired renal function and long-term mortality is well established (14). And, elevated BUN level is highly predictive of mortality, independent of creatinine in a heterogeneous critically ill population (15). In the literature a limited number of sources have shown the relationship of serum osmolality with mortality (16,17).
In the present study, we aimed to assess serum osmolality in patients with AMI; to evaluate the potential relationships between other biochemical factors and mortality rates in a six year of follow-up.

Material and Methods
A retrospective study of 204 patients with a first attack AMI, who were referred to our tertiary referral centre for angiography between the period January 2008 -June 2009 were included; informed consent was not assumed necessary because of the retrospective observational nature of the study and all steps were taken to ensure the anonymity of the data.
The study was performed in accordance with the Declaration of Helsinki and approved by the local ethics committee.
Definition of AMI is made when there is a rise /fall of cardiac biomarkers, with the evidence of symptoms, suggestive electrocardiographic changes, or imaging evidence of new loss of viable myocardium or regional abnormality of wall motion (18). To determine the prognostic factors that affect overall survival time Cox proportional hazard regression analysis with backward selection procedure was performed and results of the final step was reported. Results were reported as hazard ratios with 95% confidence intervals (CI) and related p-values.
Receiver operating characteristic curve analysis was used to assess the ability of the plasma osmolality to predict the 6-year mortality. p<0.05 was considered statistically significant. IBM SPSS v.20 was used for statistical analysis.
Age, hypertension and presence of diabetes mellitus were associated with high osmolality but smoking was not associated with osmolality (Table1).

Use of drugs containing Angiotensin Converting Enzyme (ACE)
inhibitor, acetylsalicylic acid (ASA), Statins and Oral Antidiabetic Drugs was also associated with high osmolality (Table2).
Mean follow-up time was 61.31±1.68 months. Sixteen cases died while in hospital. In total, 25 cases died within 1 year, 28 cases within 2 years, 32 cases within 3 years, 34 cases within 4 years, 36 cases within 5 years and 38 cases within 6 years.
Of the 16 deaths in the hospital, 15 were in the group with high osmolality (> 300 mOsm). Similarly, in the following years mortality was higher in the group with higher osmolality. In the high osmolality group, heart rate was higher in this group and the mean blood pressure was lower than the other groups.
In addition, high osmolality was associated with rales, stable angina pectoris presence and low ejection fraction. Number of effected vessels increased with high osmolality (Table3).
High osmolality was associated with hemogram and biochemical parameters (Table4). Fasting glucose level, BUN and sodium are the parameters we calculated the osmolality in the formula. In addition, creatinine, AST, ALT, total cholesterol, LDL and uric acid were associated with osmolality. Hemoglobin decreased with high osmolality, but WBC increased with high osmolality. Platelet count was the highest in the normal osmolality group. Sedimentation and cardiac biomarkers were high in the group with high osmolality. Hs-CRP values were more than 5-fold higher in the high osmolality group than in the other groups.
Osmolality and Hs-CRP were considered for cox regression, moderate osmolality was found protective factor and reduced death risk %87.10 (p=0.007). Hs-CRP was found as a risk factor and one unit increase in Hs-CRP will increase death risk 1.02 time (p=0.002).
Kaplan Meier analysis results showed high osmolality was associated with decreased mortality compared with low or normal osmolality (p=0.001). In addition sodium and fasting blood glucose values were also associated with mortality (p=0.031 and p<0.001 respectively) (Table5 and Figure1). Table 6 demonstrates cox regression analysis results of the patients (n=181) for mortality, excluding the patients with a history of diabetes mellitus and chronic kidney disease (n=23). BMI> 25 was defined as a risk factor for 6-year mortality. In the case of BMI> 25, the risk of mortality was 3.01 times greater.       Sodium level >145 was defined as a risk factor for 6-year mortality. If the sodium level was >145, the risk of mortality was 7.95 times more than if it was ≤145. Hypertension was defined as a risk factor for 6-year mortality. The risk of mortality was 3.16 times higher in case of hypertension.
After receiver operating characteristic curve analysis, the best cut-off value of the plasma osmolality to predict the in-

Discussion
The present study demonstrated us that osmolality increases the mortality risk in patients without a history of diabetes mellitus and chronic kidney disease. Previous studies suggested that osmolality is a risk factor for mortality (16,17,19) but the risk might be originated from concomitant renal insufficiency. We have shown that six year survival was better in patients with lower osmolality even in the absence of diabetes mellitus or chronic kidney disease.
Moderate osmolality was found as a protective factor and it resulted in a reduced death risk. In a recent study by Tatlisu  C -reactive protein has also been identified as independent risk marker for mortality also in patients with AMI (20). CRP levels predict the risk for death or AMI within 30 days among patients undergoing percutaneous coronary intervention (21)(22)(23)(24). CRP has been suggested to indicate generalized inflammation and participates directly in cardiovascular events (25). The high levels of CK-MB and troponin in the high osmolality group seem to support the idea that osmolality can be a reliable cardiac marker. Osmolality can be useful in risk estimates because it is a non-invasive method that is easy to calculate. Estimating the osmolality of sodium, BUN and glucose in routine tests will provide a cost-effective risk prediction.
Several limitations have to be considered. Data from the present study were collected in a single centre. Patients were the subjects with a cardiovascular problem and the results obtained were not suitable for making a generalization for the community.Patients' biochemical and hemogram measurements were measured only at the time of referral.
If we were able to repeat these measurements over a 6-year follow-up, we could reach more useful data. Similarly, in this study, only the mortality of patients was evaluated at 6-year follow-up. If recurrent coronary events were also observed, the relationship between osmolality and coronary artery disease could be discussed in depth.

Conclusion
The higher the osmolality, the worse the six-year survival is in patients with first episode AMI even in the absence of diabetes mellitus and chronic kidney disease. Osmolality is a cheap, noninvasive and reliable parameter to guess longterm survival in patients with coronary heart disease. We believe that hyperosmolality can be targeted in treatment and prevention efforts as well as its use when evaluating outcomes of the cardiac diseases.