Hypertension is one of the chronic diseases that clinicians can encounter the most frequently in a primary care setting; its prevalence reaches up to 31.9% in patients aged 30 years or older, 55.6% in those in their 60s, and 69.5% in those in their 70s according to the fourth Korean National Health and Nutrition Examination Survey (2009).¹⁾ It would therefore be mandatory to make an accurate measurement of blood pressure, which is essential for making a correct diagnosis of hypertension and assessing the accuracy of current prescriptions. According to the current guidelines for blood pressure monitoring, clinicians are recommended to appropriately adjust factors that may affect blood pressure measurements, such as caffeine, cigarettes, position of the arm, and the size of cuff. Moreover, they are also recommended to measure blood pressure by completely exposing the upper arm. The sleeve should not be rolled up such that it has a tourniquet effect above the blood pressure cuff.^2-7) In a Korean clinical setting where clinicians cannot afford sufficient time to examine a single patient on a fixed-time basis, however, it is not easy to measure the blood pressure in patients who have taken off their top. In particular, elderly people account for a majority of ambulatory patients who feel it difficult to collaborate with their physicians for further evaluation of hypertension in an actual clinical setting. Therefore, clinicians encounter more challenging problems when they recommend the patients to take off their top in measuring blood pressure. It is therefore a common practice that blood pressure is measured with the cuff placed over the sleeve or with the sleeve rolled. As described here, there is an inconsistency between the guidelines for blood pressure monitoring and actual clinical practice. We have therefore questioned whether there would be any differences in blood pressure measurements depending on the status of patients' wearing their top.

Holleman et al.⁸⁾ conducted a study in 36 patients, and other several English reports have shown that there is no significant correlation between blood pressure measurements and the status of wearing the top. But this has not been described in Korean literature up to present. Given the above background, we conducted this study to examine whether there are any differences in blood pressure measurements among the three different settings: the sleeve group, the rolled-sleeve group, and the bare arm group.

Baseline characteristics of the patients are summarized in Table 1. The mean age of the patients was 53.7 ± 11.8 years. Our clinical series of patients (n = 141) consisted of 117 men (83%) and 24 women (17%). There were 59 hypertensive patients (41.8%) and 28 diabetic patients (19.9%). Moreover, there were 118 patients (83.7%) where the thickness of the sleeve was smaller than 2 mm.

Following a comparison between blood pressure measurements of the first reading and those of the second reading in three different settings, it was found that there were no significant differences between the two readings (Table 2). The mean values of systolic blood pressure (SBP) were 128.5 ± 10.6 mm Hg in the sleeved group, 128.3 ± 10.8 mm Hg in the rolled sleeve group, and 128.3 ± 10.7 mm Hg in the bare arm group. These results indicate that there were no significant differences among the three groups (P = 0.32). In addition, the mean values of diastolic blood pressure (DBP) between measurements of the first reading and those of the second one were 80.7 ± 6.1 mm Hg in the sleeve group, 80.7 ± 6.1 mm Hg in the rolled sleeve group, and 80.6 ± 5.9 mm Hg in the bare arm group. These results indicate that there were no significant differences among the three groups (P = 0.77) (Table 3).

In addition, we divided the patients into subgroups based on age, sex, past or current history of hypertension or diabetes mellitus, thickness of the sleeve, weight, a drinking history, and a smoking history. There were no significant differences in SBP and DBP among the three groups among the subgroups (Tables 4, 5).

In this study, we compared blood pressure measurements obtained from three different settings in a total of 141 patients: the sleeve group, the rolled-sleeve group, and the bare arm group. As a result, there were no significant differences among the three settings. Holleman et al.⁸⁾ compared blood pressure measurements obtained using an automatic sphygmomanometer between the sleeve group and the bare arm group in 36 patients, thus reporting that there were no significant differences in SBP and DBP between the two settings. Kahan et al.¹²⁾ conducted a study using an automatic sphygmomanometer in 201 patients, who were assigned to three different groups: the sleeve group, rolled sleeve group, and the bare arm group. Then, following a comparison of blood pressure measurements using a repeatedmeasure ANOVA, it was found that there were no significant differences among the three groups. However, the authors also noted that some patients with a SBP of ≥ 140 mm Hg had a maximum difference of 32 mm Hg in the mean blood pressure measurements between the sleeve group and other groups. Thus, they proposed that blood pressure should be measured for the bare arm in hypertensive patients. In this study, we found that there were no significant differences even in hypertensive patients. We strictly followed the guidelines for blood pressure monitoring. We found no patients with a great discrepancy in blood pressure measurements, which is not in agreement with previous reports.

Liebl et al.¹³⁾ measured blood pressure using both a mercury sphygmomanometer and an automatic sphygmomanometer in 201 patients. Considering that most clothes have sleeves with a thickness of < 1 mm, the above authors enrolled only patients who wore clothed with sleeve with a thickness of < 2 mm. They measured SBP and DBP using both devices and then compared the results between the sleeve group and the bare arm group using Pearson's correlation analysis. As a result, there were no significant differences between the two different settings. Ma et al.¹⁴⁾ reported that sleeve thickness had no correlation with the difference between bare- and sleeved-arm readings. In this study, we did not restrict the clothes the patients wore when visiting Gangneung Asan Hospital. We therefore enrolled patients who wore clothes with sleeves with a thickness of > 2 mm. We found that thickness of the clothes did not affect measurement of blood pressure if a 2-cm space between the sleeve and the upper arm was made.

According to several studies, if clinicians use a cuff with an inappropriate size for the measurement of blood pressure, this will cause measurement errors.^15-19) Considering that 41.1% of patients in this study were obese, there was a limitation in that we used a cuff with consistently the same size with no respect to diameter of the arm. Moreover, was is another limitation that we did not enroll pediatric patients and failed to obtain sufficient data from younger patients. Additionally, we used an automatic sphygmomanometer rather than a classic mercury sphygmomanometer in measuring the blood pressure in the current study. This is because we attempted to minimize measurement errors because we had to measure blood pressure a total of six times in each patient. Kahan et al.¹²⁾ and Ma et al.¹⁴⁾ measured blood pressure using a commercially automatic oscillometric device. There was also a report that oscillometric blood pressure measurements showed no difference from a classic mercury sphygmomanometer.¹³⁾ We therefore presumed that the use of a classic mercury sphygmomanometer was not one of the limitations of the current study.

To summarize, our results showed that there were no significant differences in blood pressure measurements among the three different settings (the sleeve group, the rolled sleeve group, and the bare arm group). It can therefore be concluded that clothes-wearing status does not have a clinically important effect on blood pressure measurements.