Hospital-Acquired Serum Chloride Derangements and Associated In-Hospital Mortality

Background: We aimed to describe the incidence of hospital-acquired dyschloremia and its association with in-hospital mortality in general hospitalized patients. Methods: All hospitalized patients from 2009 to 2013 who had normal admission serum chloride and at least two serum chloride measurements in the hospital were studied. The normal range of serum chloride was defined as 100–108 mmol/L. Hospital serum chloride levels were grouped based on the occurrence of hospital-acquired hypochloremia and hyperchloremia. The association of hospital-acquired hypochloremia and hyperchloremia with in-hospital mortality was analyzed using logistic regression. Results: Among the total of 39,298 hospitalized patients, 59% had persistently normal hospital serum chloride levels, 21% had hospital-acquired hypochloremia only, 15% had hospital-acquired hyperchloremia only, and 5% had both hypochloremia and hyperchloremia. Compared with patients with persistently normal hospital serum chloride levels, hospital-acquired hyperchloremia only (odds ratio or OR 2.84; p < 0.001) and both hospital-acquired hypochloremia and hyperchloremia (OR 1.72; p = 0.004) were associated with increased in-hospital mortality, whereas hospital-acquired hypochloremia only was not (OR 0.91; p = 0.54). Conclusions: Approximately 40% of hospitalized patients developed serum chloride derangements. Hospital-acquired hyperchloremia, but not hypochloremia, was associated with increased in-hospital mortality.