Prognostic value of hyponatremia in hospitalized patients with worsening heart failure - insights from the outcomes of a prospective trial of intravenous milrinone for exacerbations of chronic heart failure (OPTIME - CHF)

In a post – hoc analysis we thought to describe the incidence of hyponatremia and its impact on mortality and rehospitalizations in hospitalized patients (pts.) with heart failure (HF) enrolled in the OPTIME – CHF trial. Methods: The OPTIME – CHF trial randomized 949 pts. with systolic dysfunction hospitalized for worsening HF and not requiring intravenous inotropic support to receive 48 – 72 hours of intravenous milrinone or placebo. Cox proportional hazards regression analysis was used to explore the relationship between baseline hyponatremia (Na≤136 mEq/l) and mortality at sixty days. Logistic regression was used to examine hyponatremia and rehospitalization or death at 60 days. Results: There were 256 pts. with Na≤136 mEq/l and 687 pts. with Na >136 mEq/l. The results are showed in Table 1. Conclusions: Hyponatremia (Na≤136 mEq/l) is common in hospitalized pts. with worsening HF. After adjusting for baseline variables, the Cox regression analysis showed that baseline Na, when modeled linearly, predicts an increased mortality at sixty days: Na (per 5 mEq/l) has a Hazard Ratio of 0.75 with 95% CI 0.6 - 0.95, p = 0.018. Future treatments aimed at normalizing serum sodium might improve the survival and should be evaluated in randomized trials.1Baseline characteristics and clinical outcomes in patients hospitalized with heart failure from the OPTIME in CHF TrialVariableBaseline Na≤136 mEq/l - 256 pts. (27%)Baseline Na >136 mEq/l - 687 pts. (73%)p valueAge (years) ∗Median, 25th, 75th67 (53, 76)68 (57, 76)0.669Male sex (%)68660.577Ischemic etiology of HF (%)55500.18NYHA Class III - IV (%)96920.047Elevated JVP (%)72660.1157Pulmonary rales (%)83810.444Baseline HR (bpm)∗Median, 25th, 75th84 (72, 96)84 (72, 96)0.5654Baseline SBP (mmHg) ∗Median, 25th, 75th114 (100, 130)121 (108, 135)0.0001Baseline DBP (mm Hg) ∗Median, 25th, 75th70 (60, 78)70 (62, 80)0.015Baseline BUN (mg/dl) ∗Median, 25th, 75th11.8 (7.1, 17.5)8.6 (6.1, 13.2)0.0001Baseline Cr (mg/dl) ∗Median, 25th, 75th1.4 (1.1, 1.8)1.3 (1.1, 1.7)0.0172Mortality at 60 days (%)16%7%0.0001Log – rank statisticRehospitalizations/ death at 60 days (%)42%33%0.017∗ Median, 25th, 75th∗∗ Log – rank statistic Open table in a new tab In a post – hoc analysis we thought to describe the incidence of hyponatremia and its impact on mortality and rehospitalizations in hospitalized patients (pts.) with heart failure (HF) enrolled in the OPTIME – CHF trial. Methods: The OPTIME – CHF trial randomized 949 pts. with systolic dysfunction hospitalized for worsening HF and not requiring intravenous inotropic support to receive 48 – 72 hours of intravenous milrinone or placebo. Cox proportional hazards regression analysis was used to explore the relationship between baseline hyponatremia (Na≤136 mEq/l) and mortality at sixty days. Logistic regression was used to examine hyponatremia and rehospitalization or death at 60 days. Results: There were 256 pts. with Na≤136 mEq/l and 687 pts. with Na >136 mEq/l. The results are showed in Table 1. Conclusions: Hyponatremia (Na≤136 mEq/l) is common in hospitalized pts. with worsening HF. After adjusting for baseline variables, the Cox regression analysis showed that baseline Na, when modeled linearly, predicts an increased mortality at sixty days: Na (per 5 mEq/l) has a Hazard Ratio of 0.75 with 95% CI 0.6 - 0.95, p = 0.018. Future treatments aimed at normalizing serum sodium might improve the survival and should be evaluated in randomized trials.