Variability in Metabolic Parameters: A Prognostic Surrogate Marker for MI, Stroke and Death? Associations of Variability in Blood Pressure, Glucose and Cholesterol Concentrations, and Body Mass Index With Mortality and Cardiovascular Outcomes in the General Population

A study by Mee Kyoung Kim and her colleagues published in Circulation has shown that there is a graded association between the number of high variability parameters like fasting blood glucose and total cholesterol levels, systolic blood pressure, and body mass index and cardiovascular outcomes. They showed that the mentioned variables may be considered as independent predictors of mortality and cardiovascular events.

It was thought that the most important factor for predicting health outcomes can be estimated by the means of important metabolic and physiologic parameters. Kim and her colleagues demonstrated that variations and fluctuations in the levels of the following parameters may be associated as independent factors for predicting health outcomes: Blood pressure (BP), fasting blood glucose (FBG), total cholesterol levels (TC), and body mass index (BMI).

Investigators performed a nationwide population-based cohort study using The Korean National Health Insurance System (NHIS) dataset which contains more than 50 million entries on population health information as the primary data. The study population included 6748773 people. They included those who had recent health examination in the past 3 years before the study begins (the index period) and had been through more than 2 health examination visits during the past 3 years before the index period. They also excluded patients with previously diagnosed and treated diabetes mellitus, hypertension, dyslipidemia as these treatments can affect body weight and other measurements, and to exclude the possible effect of variable compliance with drug treatment. They also excluded people with a preexisting history of myocardial infarction (MI) or stroke to avoid confounding by preexisting diseases and to minimize the possible effects of reverse causality. The investigators developed a scoring system for variability among all 4 variables. They defined the endpoints of the study as newly diagnosed MI, stroke, or death. They followed up the patients for 5.5 years.

The investigators used unadjusted Kaplan–Meier curves for analyzing the cumulative incidence of primary outcomes according to the number of parameters with high variability. They also used the hazard ratio (HR) and 95% CI for analyzing all-cause mortality, MI, and stroke using the Cox proportional-hazards model. They found there is a linear relationship between the number of high-variability parameters and related to the outcome measures, as the group with high variability for all 4 parameters had a significantly higher risk for all-cause mortality, MI, and stroke.

For the highest quartile in FBG variability compared with the lowest quartile, the risk of all-cause mortality increased by 20%, MI by 16%, and stroke by 13%. For the highest quartile in TC variability compared with the lowest quartile, the risk of all-cause mortality increased by 31%, MI by 10%, and stroke by 6%. For the highest quartile in systolic BP variability compared with the lowest quartile, the risk of all-cause mortality increased by 19%, MI by 7%, and stroke by 14%. For the highest quartile in BMI variability compared with the lowest quartile, the risk of all-cause mortality increased by 53%, MI by 14%, and stroke by 14%. Kim found that there was a graded association between high variability in FBG and TC concentrations, SBP, and BMI and higher risks for mortality, MI, and stroke development during a 5.5-year follow-up period. They also found that variability in BMI was an independent predictor of the outcomes studied. Their study confirmed a reverse association between the change in body weight and all-cause mortality, which include both weight loss and weight gain. They also noticed that increased glucose variability is related to increased mortality in people. This is shown also by other studies to be regardless of the patients having diabetes mellitus or not. The association was shown to be particularly strong among critically ill patients in the euglycemic range.

The investigators addressed the limitations of the study including the observational method of the study which limited them to have the conclusion of causality for the associations they found between variability and end points. They also acknowledged that they had a limitation regarding intentional versus unintentional BMI variability. Kim et al. remarked, “Our work states there is an association between high variability in metabolic parameters and adverse health outcomes. This association is not only limited to diseased populations but also exist in relatively healthy populations, although the mechanism could be somewhat different. Indeed, variability in metabolic parameters may be a prognostic surrogate marker for predicting mortality and cardiovascular outcomes.”