According to epidemiological studies, researchers observed and examined that an increased PLT count is associated with an increased incidence of coronary heart disease and all-cause mortality. They observed a significant inverse association between indoor cold exposure and PLT count.

Platelet (PLT) count is an important determinant of PLT aggregation. In vitro, whole blood aggregation is associated with PLT count not only among healthy participants but also among patients with ischemic heart disease. In addition, antiplatelet therapy significantly decreased events of thrombotic diseases, such as acute myocardial infarction and stroke.

However, the quantification of cold exposure using outdoor temperatures is not sufficient among elderly individuals, a high-risk group for excess mortality during winter, because they spend the majority of their time (>85%) at home.  Indeed, ambulatory blood pressures among elderly individuals were more strongly associated with indoor temperature than outdoor temperature.

To our knowledge, this is the first epidemiologic study investigating the association between indoor cold exposure and PLT count in a real-life setting. The association remained significant after adjustment for potential confounders, including age, gender, hematocrit, outdoor temperature, day length, time spent out of home, smoking, ethanol intake, body weight, eGFR, antihypertensive medication, diabetes, and socioeconomic status.

 According to the categorical analysis by tertile groups based on daytime indoor temperatures, an average of 8.4 °C lower daytime indoor temperature was associated with a 5.2% higher PLT count. The present study had several limitations. First, only PLT count was assessed as a hemostatic factor. Second, we cannot determine the direction of causality between indoor temperature and PLT count from the cross-sectional analysis.

As an index of indoor cold exposure at measurement of PLT count, we used the mean value of indoor temperatures in the following day from rising time to bedtime, which may be acceptable based on high repeatability of indoor temperatures on two consecutive daytimes among 923 participants. A previous study showed that PLT count collected with 1-min and 3-min stasis showed acceptable correlation with PLT count collected without stasis (r > 0.95).

Fourth, we did not quantify the amount of clothing worn. In a previous randomized controlled trial, we found a significant increase in the amount of clothing worn due to a 10 °C-lower indoor temperature. The influence of indoor temperature on PLT count may be underestimated if the participants in cold indoor temperature wear more clothing than in the present study.

Finally, participants were not randomly selected but were volunteers. This may limit the generalization of the findings of the present study to the elderly population as a whole. However, patient characteristics, such as body mass index and eGFR, were similar to those reported in a nationwide survey of randomly selected participants.

In conclusion, researchers demonstrate a significant inverse association between PLT count and daytime indoor temperature in the homes of elderly individuals during winter.