Kohei Inukai, Mikito Takefuji, Shingo Kazama, Ryota Ito, Hiroaki Hiraiwa, Toru Kondo, Ryota Morimoto, Shouji Matsushima, Keisuke Kida, Toshiharu Ninomiya, Nobuyuki Enzan, Masataka Ikeda, Takeshi Kitai, Tatsunori Taniguchi, Takeshi Tohyama, Hiroyuki Tsutsui, Yuya Matsue, Toyoaki Murohara, Takahiro Okumura
Hypochloraemia identified patients with heart failure at substantially increased cardiovascular risk independent of established prognostic markers. Exploratory analyses suggested a potential association between circulating UCN2 and chloride homeostasis that warrants mechanistic investigation.
BACKGROUND: Hypochloraemia has emerged as a prognostic marker in heart failure, although the mechanisms underlying chloride dysregulation remain incompletely understood. We evaluated the prognostic significance of serum chloride levels and explored clinical and biochemical correlates of hypochloraemia including circulating urocortin 2 (UCN2).
METHODS: This multicentre prospective cohort study included 3941 patients discharged after hospitalisation for acute or worsening heart failure with available serum chloride measurements. Patients were categorised according to discharge chloride levels as hypochloraemia, normochloraemia or hyperchloraemia. Associations between chloride status and 2-year outcomes were assessed using multivariable Cox proportional hazards and competing risk models. Exploratory analyses of circulating UCN2 concentrations were performed in a predefined subset of 499 patients.
RESULTS: Hypochloraemia was independently associated with adverse outcomes after multivariable adjustment, including all-cause mortality (adjusted HR, 2.09; 95% CI 1.64 to 2.67; p<0.001), cardiovascular mortality (adjusted HR, 2.29; 95% CI 1.64 to 3.20; p<0.001), heart failure rehospitalisation (adjusted HR, 1.30; 95% CI 1.02 to 1.67; p=0.04) and the composite outcome (adjusted HR, 1.46; 95% CI 1.20 to 1.78; p<0.001). Competing risk analyses demonstrated that the excess mortality risk associated with hypochloraemia was predominantly cardiovascular. In models including both serum chloride and sodium, chloride remained independently associated with all-cause mortality whereas sodium did not. In exploratory analyses, circulating UCN2 concentrations showed a modest inverse association with serum chloride levels after adjustment for sodium, renal function, nutritional status and clinical severity (B=-0.37; 95% CI -0.72 to -0.02; p=0.04).
CONCLUSIONS: Hypochloraemia identified patients with heart failure at substantially increased cardiovascular risk independent of established prognostic markers. Exploratory analyses suggested a potential association between circulating UCN2 and chloride homeostasis that warrants mechanistic investigation.