Platelet distribution width (PDW) is a quantitative measure of variability in platelet size. It has been hypothesized that increased PDW may be a predictor of mortality. However, this hypothesis has not been examined in a community‐based population.
Kaplan‐Meier survival curves of participants in the lowest (first) and uppermost (fourth) quartiles of platelet distribution width.
We used the Third National Health and Nutrition Examination Survey to examine association between PDW and all‐cause mortality and cardiovascular (CV) mortality. We used the Cox proportional hazards model to calculate hazards ratios (HRs) and 95% confidence intervals (CIs) in 14,876 participants who were older than 20 years and were free of clinical CV disease at baseline.
There were 3377 all‐cause deaths and 1397 CV deaths during the follow‐up (median follow‐up, 16.4 years; range, 0–18 years). The mean (SD) age of the participants was 47 (19) years, 54% were female, 40% were white, 26% were current smokers, and 25% had a history of hypertension. Mean (SD) PDW was 16.4% (0.5). Univariate analyses found a significant association between PDW and all‐cause and CV mortality (HR, 1.48 per unit increase in PDW; 95% CI, 1.31–1.67; and HR, 1.51 per unit increase in PDW; 95% CI, 1.31–1.75, respectively). This association remained statistically significant in multivariable analyses after adjusting for age, sex, race, C‐reactive protein, diabetes, hypertension, smoking status, glomerular filtration rate, and total cholesterol (HR for all‐cause mortality, 1.30; 95% CI, 1.19–1.42; HR for CV mortality, 1.27; 95% CI, 1.12–1.43). Moreover, compared with participants who were in the lowest quartile of PDW (mean, 15.8%), those in the upper quartile of PDW (mean,17.2%) had 47% (HR, 1.47; 95% CI, 1.19–1.83) and 34% (HR, 1.34; 95% CI, 1.04–1.72) greater hazards of all‐cause death and CV death, respectively.
PDW is an independent predictor of all‐cause and CV mortality and may identify high‐risk individuals. We propose that this test should be routinely reported with other hematological parameters in complete blood count reports.
R. Qayyum ‐ none; J. Adomaityte ‐ none; M. Khawaja ‐ none; A. Rehman ‐none; S. Shakeel ‐ none; M. Amer ‐ none