SERUM CHLORIDE AND ADVERSE CARDIOVASCULAR RISK IN THE WEST OF SCOTLAND GENERAL POPULATION

Abstract

Objective: To examine the association between serum Cl- and cardiovascular and non-cardiovascular morbidity and mortality outcomes in the West of Scotland general population. Design and method: Patients aged > 18 years resident in NHS Greater Glasgow and Clyde who had serum Cl- measured at baseline were included in the study. The study was approved by NHSGGC Local Privacy Advisory Committee (LPAC) (GSH/16/CA/012). Baseline defined as: first serum Cl- taken outwith one-month of an acute hospital admission or one-month post discharge and in the absence of prescription for diuretic therapy in the previous 3 months or serum sodium < =125 or > = 150 mmol/L; serum creatinine > 125 micromol/L. ICD-10 codes were used to characterise diagnosis and cause of death. Primary outcomes were a composite of fatal/non-fatal events for each specified outcome: all CV events, MI, IHD, heart failure, all stroke, stroke subtypes, PVD, cardiac dysrhythmia, COPD, asthma pneumonia/influenza, chronic liver disease, diabetes, kidney disease. The study population were divided into ten groups based on serum Cl- (< = 99,100–101.9,102–102.9,103–103.9,104–104.9,105–105.9,106–106.9,107–107.9,108–110.9 > = 111mmol/L). Outcomes assessed using Cox-PH models. Model 1 was adjusted for age and sex. Model 2–6 for age, sex, and variables including prevalent cardiovascular disease, prevalent diabetes, Scottish Indices of Multiple Deprivation (SIMD), serum Na+, serum bicarbonate, total cholesterol. Results: The study population contained 519,957 patients with serum Cl- measured at baseline. The population were 50.6 (17.7) years old; 43.5% were male; 9% had diabetes at baseline and 3.8% had a history of cardiovascular disease. One third (33.4%) of the population resided in an area within the SIMD quintile 1. There was a ’U’-shaped non-linear association between serum Cl- and risk of fatal/non-fatal events. The results from Model 1 are presented in Figure 1. The greatest risk was seen in groups 1–3 (Cl- < 103mmol/L) and groups 8–10 (Cl- > 106mmol/L) compared with group 6 (Cl- 104–105mmol/L) for all CV events, MI, heart failure, PVD, all stroke, ischaemic stroke, COPD, pneumonia/influenza, liver disease and kidney disease. Conclusions: In a large general population from the West of Scotland, there was a ’U’-shaped non-linear association between serum Cl- and risk of fatal/non-fatal CV and non-CV events.