Associations between PM2.5 metal components and QT interval length in the Normative Aging Study

Abstract

Background: Several studies have found associations between increases in QT interval length, a marker of cardiac electrical instability, and short-term fine particulate matter (PM_2.5) exposures. To our knowledge, this is the first study to examine the association between specific PM_2.5 metal components and QT interval length.

Methods: We measured heart-rate corrected QT interval (QTc) duration among 630 participants in the Normative Aging Study (NAS) based in Eastern Massachusetts between 2000 and 2011. We utilized time-varying linear mixed-effects regressions with a random intercept for each participant to analyze associations between QTc interval and moving averages (0-7 day moving averages) of 24-h mean concentrations of PM_2.5 metal components (vanadium, nickel, copper, zinc and lead) measured at the Harvard Supersite monitoring station. Models were adjusted for daily PM_2.5 mass estimated at a 1 km × 1 km grid cell from a previously validated prediction model and other covariates. Bayesian kernel machine regression (BKMR) was utilized to assess the overall joint effect of the PM_2.5 metal components.

Results: We found consistent results with higher lead (Pb) associated with significant higher QTc intervals for both the multi-pollutant and the two pollutant (PM_2.5 mass and a PM_2.5 component) models across the moving averages. The greatest effect of lead on QTc interval was detected for the 4-day moving average lead exposure. In the multi-pollutant model, each 2.72 ng/m³ increase in daily lead levels for a 4-day moving average was associated with a 7.91 ms (95% CI: 3.63, 12.18) increase in QTc interval. In the two-pollutant models with PM_2.5 mass and lead, each 2.72 ng/m³ increase in daily lead levels for a 4-day moving average was associated with an 8.50 ms (95% CI: 4.59, 12.41) increase in QTc interval. We found that 4-day moving average of copper has a negative association with QTc interval when compared to the other PM_2.5 metal components. In the multi-pollutant model, each 1.81 ng/m³ increase in daily copper levels for a 4-day moving average was associated with an -3.89 ms (95% CI: -6.98, -0.79) increase in QTc interval. Copper's essential function inside the human body could mediate its cardiotoxicity on cardiac conductivity and explain why we found that copper in comparison to the other metals was less harmful for QTc interval.

Conclusions: Exposure to metals contained in PM_2.5 are associated with acute changes in ventricular repolarization as indicated by QT interval characteristics.

Keywords: Air pollution; Fine particulate matter; Metals; QT interval; Ventricular repolarization.

Fig. 1.

Changes in milliseconds and 95% CI in QTc interval for an IQR increase in zero to seven day moving average of each PM_2.5 metal component. The results are presented in a multi-pollutant model where all the metal components are included in the same model, the two-pollutant model which includes each individual PM_2.5 metal component and PM_2.5 mass and the BKMR model. The results for BKMR are reported with the 95% posterior credible interval (PCI) with the other exposures fixed at their 50th percentile. All models are adjusted for PM_2.5 mass, daily ozone, age (years), race, maximum years of education, BMI (kg/m2), total cholesterol (mg/dL), mean arterial pressure (mmHg), diabetic status, use of beta blocker medication, alcohol intake (2 or more drinks per day or less than 2 drinks per day as reference), smoking status (current, former or never as reference), census tract percent of population age 25 years or older with less than a high school diploma, air temperature (°C), relative humidity (%) and seasonality (sine and cosine).

Fig. 2.

Overall joint effect of the PM_2.5 metal mixture for 0–7 day moving averages with QTc interval length estimated by Bayesian Kernel Machine Regression (BKMR). This figure compares the estimated change in QTc interval length when all predictors are at a certain quantile with the value when all of them are at their 50th percentile. BKMR models were adjusted for PM_2.5 mass, daily ozone, age (years), race, maximum years of education, BMI (kg/m2), total cholesterol (mg/dL), mean arterial pressure (mmHg), diabetic status, use of beta blocker medication, alcohol intake (2 or more drinks per day or less than 2 drinks per day as reference), smoking status (current, former or never as reference), census tract percent of population age 25 years or older with less than a high school diploma, air temperature (°C), relative humidity (%) and seasonality (sine and cosine).

Fig. 3.

Change in QTc interval length and 95% CI for an IQR increase in 0–7 day moving average of each PM^2.5 metal component in the multi-pollutant model stratified by season. Model was adjusted for PM_2.5 mass, daily ozone, age (years), race, maximum years of education, BMI (kg/m2), total cholesterol (mg/dL), mean arterial pressure (mmHg), diabetic status, use of beta blocker medication, alcohol intake (2 or more drinks per day or less than 2 drinks per day as reference), smoking status (current, former or never as reference), census tract percent of population age 25 years or older with less than a high school diploma, air temperature (°C), relative humidity (%) and seasonality (sine and cosine).

Fig. 4.

Changes in milliseconds and 95% CI in QTc interval for an IQR increase in zero to seven day moving average of each PM_2.5 metal component. The results are presented in a multi-pollutant model where all the metal components are included in the same model except for the indicated PM_2.5 metalcomponent. The models were adjusted for PM_2.5 mass, daily ozone, age (years), race, maximum years of education, BMI (kg/m2), total cholesterol (mg/dL), mean arterial pressure (mmHg), diabetic status, use of beta blocker medication, alcohol intake (2 or more drinks per day or less than 2 drinks per day as reference), smoking status (current, former or never as reference), census tract percent of population age 25 years or older with less than a high school diploma, air temperature (°C), relative humidity (%) and seasonality (sine and cosine).