Prognostic Importance of Hyponatremia in Acute ST-Elevation Myocardial Infarction Alexander Goldberg, MD, Haim Hammerman, MD, Sirouch Petcherski, MD, Alexander Zdorovyak, MD, Sergey Yalonetsky, MD, Michael Kapeliovich, MD, PhD, Yoram Agmon, MD, Walter Markiewicz, MD, Doron Aronson, MD PURPOSE: To determine the prevalence and prognostic im- plications of hyponatremia in the setting of acute ST-elevation myocardial infarction. METHODS: The study sample consisted of 1047 consecutive patients presenting with acute ST-elevation myocardial infarc- tion. Plasma sodium concentrations were obtained on admis- sion and at 24, 48, and 72 hours thereafter. Infarct size was determined by echocardiographic examination that was per- formed on day 2 or 3 of hospitalization. RESULTS: Hyponatremia, defined as a plasma sodium level 135 mmol/L (135 mEq/L), was present on admission in 131 patients (12.5%) and developed during the first 72 hours of hospitalization in 208 patients (19.9%). Plasma sodium levels decreased to 130 mmol/L in 45 patients (4.3%). In a multi- variate logistic regression analysis, hyponatremia was indepen- dently associated with 30-day mortality. The risk of 30-day mortality associated with hyponatremia on admission (odds ra- tio [OR] = 2.0; 95% confidence interval [CI]: 1.0 to 3.9; P = 0.04) was similar to that of hyponatremia developing after ad- mission (OR = 2.4; 95% CI: 1.5 to 4.2; P = 0.002). The risk of 30-day mortality increased with the severity of hyponatremia, with an odds ratio of 2.1 in patients with sodium levels between 130 and 134 mmol/L (95% CI: 1.2 to 3.5; P = 0.007) and 3.4 in those with levels 130 mmol/L (95% CI: 1.5 to 7.8; P = 0.002). CONCLUSION: Hyponatremia on admission or early devel- opment of hyponatremia in patients with acute ST-elevation myocardial infarction is an independent predictor of 30-day mortality, and prognosis worsens with the severity of hypona- tremia. Further studies are required to determine if plasma so- dium levels may serve as a simple marker to identify patients at high risk. Am J Med. 2004;117:242–248. ©2004 by Elsevier Inc. H yponatremia is a common electrolyte disorder among hospitalized patients (1), especially in the postoperative period (2) and in patients with heart failure, nephrotic syndrome, or cirrhosis (3,4). It is recognized as a predictor of adverse outcomes in hospi- talized patients, and its prognostic implications are usu- ally attributed to the severity of the underlying condition (1,5). Hyponatremia has been shown to be a predictor of cardiovascular mortality among patients with heart failure (6,7). In these patients, hyponatremia has been related to the nonosmotic release of vasopressin (8), ac- tivation of the renin-angiotensin system (6,9), and cate- cholamine production (10). In fact, the neurohormonal activation that accompanies acute myocardial infarction is similar to that which accompanies heart failure (11). However, while the prognostic value of hyponatremia in chronic heart failure is well established (6), data on the prevalence and prognostic importance of hyponatremia in the setting of acute myocardial infarction are lacking (12). Therefore, we sought to study the prevalence of hy- ponatremia during the acute phase of ST-elevation myo- cardial infarction and to determine its usefulness in pre- dicting short-term survival. METHODS Patients The study sample comprised 1047 consecutive patients admitted with acute ST-elevation myocardial infarction to Rambam Medical Center, Haifa, Israel, between July 2000 and March 2003. The study was approved by the investigational review committee on human research. Myocardial infarction was diagnosed in the presence of chest pain lasting 20 minutes, diagnostic serial electro- cardiographic (ECG) changes with characteristic ECG al- terations consisting of new pathological Q waves or ST- segment and T-wave changes, and plasma creatine kinase–MB elevation greater than twice the normal levels or elevated cardiac troponin T levels (13). Qualifying pa- tients received thrombolytic therapy (tissue-type plas- minogen activator or streptokinase) or underwent pri- mary coronary angioplasty according to the discretion of the attending cardiologist. Creatine kinase levels were de- termined at presentation and subsequently at 4-hour in- tervals until reaching a peak. Assessment of left ventricu- lar ejection fraction by echocardiography was performed either on day 2 or 3 of hospitalization in most patients or earlier if clinically indicated. From the Department of Cardiology, Rambam Medical Center (AG, HH, SP, AZ, SY, MK, YA, WM, DA), and Rappaport Medical School (HH, WM), Haifa, Israel. Requests for reprints should be addressed to Doron Aronson, MD, Department of Cardiology, Rambam Medical Center, Bat Galim, POB 9602, Haifa 31096, Israel, or d_aronson@rambam.health.gov.il. Manuscript submitted November 12, 2003, and accepted in revised form March 9, 2004. 242 © 2004 by Elsevier Inc. 0002-9343/04/$–see front matter All rights reserved. doi:10.1016/j.amjmed.2004.03.022