Community-based serum chloride abnormalities predict mortality risk

PLOS ONE RESEARCH ARTICLE Community-based serum chloride abnormalities predict mortality risk Tali Shafat1,2, Victor Novack1, Leonid Barski2, Yosef S. Haviv ID3* 1 Clinical Research Center, Soroka University Medical Center and the Faculty of Health Sciences, BenGurion University of the Negev, Beer Sheba, Israel, 2 Department of Internal Medicine F, Soroka University Medical Center and the Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheba, Israel, 3 Nephrology Department, Soroka University Medical Center and the Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer Sheba, Israel a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * Abstract Introduction OPEN ACCESS Citation: Shafat T, Novack V, Barski L, Haviv YS (2023) Community-based serum chloride abnormalities predict mortality risk. PLoS ONE 18(2): e0279837. https://doi.org/10.1371/journal. pone.0279837 Editor: Tatsuo Shimosawa, International University of Health and Welfare, School of Medicine, JAPAN Received: June 21, 2021 Accepted: December 15, 2022 Published: February 21, 2023 Copyright: © 2023 Shafat et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting information files. This population-based study aimed to investigate the prognostic value of ambulatory serum chloride abnormalities, often ignored by physicians. Methods The study population included all non-hospitalized adult patients, insured by "Clalit" Health Services in Israel’s southern district, who underwent at least 3 serum chloride tests in community-based clinics during 2005–2016. For each patient, each period with low (�97 mmol/ l), high (�107 mmol/l) or normal chloride levels were recorded. A Cox proportional hazards model was used to estimate the mortality risk of hypochloremia and hyperchloremia periods. Results 664,253 serum chloride tests from 105,655 subjects were analyzed. During a median follow up of 10.8 years, 11,694 patients died. Hypochloremia (� 97 mmol/l) was independently associated with elevated all-cause mortality risk after adjusting for age, co-morbidities, hyponatremia and eGFR (HR 2.41, 95%CI 2.16–2.69, p<0.001). Crude hyperchloremia (�107 mmol/L) was not associated with all-cause mortality (HR 1.03, 95%CI 0.98–1.09 p = 0.231); as opposed to hyperchloremia �108 mmol/l (HR 1.14, 95%CI 1.06–1.21 p<0.001). Secondary analysis revealed a dose-dependent elevated mortality risk for chloride levels of 105 mmol/l and below, well within the "normal" range. Funding: The author(s) received no specific funding for this work. Conclusion Competing interests: The authors have declared that no competing interests exist. In the outpatient setting, hypochloremia is independently associated with an increased mortality risk. This risk is dose-dependent where the lower the chloride level, the higher is the risk. Abbreviations: AG, anion gap; BB, beta blockers; CCB, calcium channel blockers; CHF, congestive heart disease; CKD, chronic kidney disease; ED, emergency department; ESRD, end stage renal PLOS ONE | https://doi.org/10.1371/journal.pone.0279837 February 21, 2023 1 / 13 PLOS ONE disease; FU, follow up; GEE, generalized estimating equation; GFR, glomerular filtration rate; HMO, health maintenance organization; HR, hazard ratio; ICU, intensive care unit; IQR, interquartile range; LOWESS, Locally Weighted Scatterplot Smoothing; MDRD, Modification of Diet in Renal Disease; MM, multiple myeloma; Mmol, milli mo; RAS, reninangiotensin- aldosterone system; SD, standard deviation; SUMC, Soroka University Medical Center; TPN, Total parenteral nutrition. Chloride and mortality Introduction Chloride is the major extracellular anion playing a key role in acid-base balance and body fluid homeostasis, distinctly from sodium. Nevertheless, serum chloride levels are often ignored despite their availability in routine chemistry blood tests. Recently, several studies among chronic heart failure patients indicated that chronic hypochloremia can serve as a prognostic marker for all-cause mortality, even better than hyponatremia [1–6]. Similar associations between hypochloremia and all-cause mortality and cardiovascular events were also found among chronic kidney disease (CKD) patients and hypertensive patients [7–9]. However, not all studies reported a methodology for the adjustment for concomitant hyponatremia, a known dominant risk factor for mortality, and except for 2 studies [1, 10], all studies examined specific patient populations rather than the general population [2–9]. The first study evaluating the prognostic value of serum chloride levels in the general outpatient population originated as an unexpected finding of a nutrition study conducted between 1979 to 1985 [1]. The second study from Mayo clinic examined a large cohort and found a U-shaped association between serum chloride levels before discharge and 1- year outcome where optimal serum chloride levels were 100–108 mmol/l [10]. Different values for the normal ranges for serum chloride have been previously reported: 100–108 mmol/L, 97–110 mmol/L or 98–106 [11, 12, 19]. It had been suggested that chloride <100 mmol/L or <102.8 is associated with higher mortality [3, 7] and in another large prospective study the lowest and highest chloride level quartiles (<104 mmol/L and >108 mmol/ L respectively) were both associated with higher all-cause mortality risk and CV events [9]. In addition, while hyperchloremia confers a mortality risk in acutely ill ICU patient [13– 16], it is currently unknown whether chronic hyperchloremia can serve as independent prognostic marker. A recent study among CKD patients’ cohorts showed that a serum chloride level of 106–108 is associated with the lowest mortality risk [8]. Finally, because in the primary care settings blood gases measurement is commonly not available, we hypothesized that community-based serum chloride test can serve as a simple and accessible proxy for blood pH [17, 18]. Taken together, the drivers to study the prognostic value of serum chloride were i. to generate a large updated population-based database from the general population to evaluate the risk associated with serum chloride abnormalities. ii. to relate the acid base status with the serum chloride level. This population-based study aimed to investigate these 2 parameters and is the first hospital-excluding study designed to characterize the significance of serum chloride levels in the general community population. Materials and methods Study population This population-based study included all adult patients, insured by "Clalit" health plan, Israel southern district, who underwent at least three serum chloride tests during 2002–2016. Clalit is an extensive (...truncated)