Admission hypochloremia predicts prolonged hospital stay and adverse in-hospital outcomes in congestive heart failure: A quantile regression analysis of 3,454 patients
A G Pradeep, A Zayed, G Rababah, A Deutsch, J Abou Chehade, B Abofrekha, H Itani, O Khayat, J Javor, C Sison, S El-SayeghAbstract
Background/Introduction
Acute Congestive heart failure (CHF) remains a leading cause of hospitalization and mortality worldwide. Despite advances in treatment, the identification of simple, cost-effective prognostic biomarkers at admission remains an unmet need. While serum chloride has proven to be an important prognostic marker in acute heart failure, most previous studies have focused on mortality and readmission outcomes. The impact of admission plasma chloride levels on hospital length of stay (LOS) and short-term in-hospital outcomes has not been clearly defined.
Purpose
This study is the first to investigate whether admission hypochloremia is associated with longer hospital stay and adverse in-hospital outcomes in acute CHF. We hypothesized that low plasma chloride would predict prolonged LOS and that this relationship would be influenced by renal function, serum sodium, and loop diuretic use.
Methods
A retrospective cohort of 3,454 adult acute CHF admissions (2020–2025) at a tertiary center was analyzed. Admission plasma chloride was grouped into five categories: <90, 90–94, 95–97, 98–104 (reference), and >104 mmol/L. The primary outcome was hospital LOS, evaluated using quantile (median) regression, which is robust to non-normal data distribution. Covariates included acute kidney injury (AKI), diabetes, chronic kidney disease (CKD), serum sodium, potassium, creatinine, pro-BNP, and loop diuretic use. Secondary outcomes were in-hospital mortality, ICU admission, need for mechanical ventilation, and 30-day readmission. Interaction terms (chloride × AKI, chloride × sodium, chloride × diuretic use) were examined to assess effect modification.
Results
Median plasma chloride was 99 mmol/L (IQR 96–102). Patients with hypochloremia (<98 mmol/L) had longer median LOS compared with those in the normal range (8.9 vs. 7.0 days; p<0.001). In adjusted quantile regression, chloride <90 and 90–94 mmol/L were associated with +2.43 days (p=0.0249) and +2.07 days (p=0.0005) longer LOS, respectively. AKI (p<0.001) and higher pro-BNP (p=0.0008) were also independent predictors of prolonged stay. Hypochloremia correlated with higher in-hospital mortality (16.2% vs. 6.7%), ICU admission (59.2% vs. 35.3%), and mechanical ventilation (22.5% vs. 11.6%), all p<0.01, but not with 30-day readmission (p=0.57). Adverse impact of hypochloremia on LOS was greater in patients with AKI, low sodium (< 135 mmol/L), or loop diuretic exposure.
Conclusion
Admission hypochloremia is a strong, independent predictor of prolonged hospital stay and adverse in-hospital outcomes in acute CHF. Given its low cost and universal availability, admission plasma chloride should be considered in risk stratification models and may help guide individualized inpatient management. Prospective studies are warranted to determine whether early hypochloremia based treatment strategies can shorten hospitalization and improve outcomes.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.