The primary aim of the project was to compare clinic BP, AMBP, cognition and HRV between healthy normal control subjects, WCHT, and BLH subjects.
Existing evidence shows that individuals with WCHT may develop persistent hypertension over a relatively short period of time . These findings were based on the prognostic significance of ABPM in comparison with clinic BP measurement . We report a significantly higher day-time ABPM (systolic and diastolic) values among WCHT, and more so among BLH subjects as compared to normal controls. The most important difference was seen in a sharp fall of BP (systolic and diastolic) between clinic-, day-time-, and night-time- BP readings among the WCHT, and though slightly less, among BLH subjects. Our results are consistent with available data in the literature , and may, in fact, correspond to the classical diagnostic findings of BP levels among WCHT subjects i.e. mean day-time BP of < 135/85 mm Hg, but > 140/90 mm Hg in the clinic according to the definition of the British Hypertension Society . However, our data cannot confirm a foreseeable transition between WCHT to sustained hypertension. Generally, the BP observed between the groups served as a means of categorizing participants on the severity of hypertension thereby enabling comparisons to be made. The comparison of clinic and 24-hour BP has proved that WCHT was apparent in this sample. It also suggests that borderline subjects have higher sustained mean BPs over 24-hour. However, our study did not address, and therefore cannot explain longitudinal effects of these measurements. Hence, whether other factors could identify the future prognosis of these variables in patients with hypertension may rely on the evidence found when cognitive function and HRV were assessed.
Five out of the nine CANTAB computer tests revealed significant differences in performance between the three groups (post-hoc). It was evident from the mean scores that WCHT poorly performed in memory tests, whereas borderline subjects had slower reaction time, but both groups performed significantly worse than normotensives. These results were consistent with the observations of Elias and colleagues . In contrast, other studies , have shown no relationship between BP and cognitive performance. These conflicting results may reflect several methodological differences in these studies. It was thought that the negative findings in the Farmer's study  was most likely due to BP measurements taken simultaneously with neuropsychological testing, or probably too few measurements were taken. Our findings agree with Elias et al of BP being inversely related to cognitive function. Despite the relatively small sample size, our results indicated a significant lack of performance in cognitive tests by subjects with WCHT and borderline when compared to controls.
The state anxiety inventory revealed that subjects with WCHT and BLH had significantly higher anxiety levels than normotensives. This may well account for this group's lack of performance in most of the cognitive tests and could clinically be the cause of the high BPs measured at the hospital i.e. WCHT. Psychological factors may be closely related to the development of hypertension . However, all BP measurements and CANTAB computer test results remained significant even when data was stratified for sex, age and anxiety scores. Thus, BPs obtained and the computer test scores were independent of differences in age, sex and anxiety levels between the three groups. It can be speculated that these intermediate groups (WCHT and borderline subjects) will develop sustained hypertension associated with poor cognitive performance and high levels of anxiety. Such findings raise the possibility that controlling anxiety levels with these variables may reduce stress and consequently lower BP. Previous work  has suggested that relaxation training can increase parasympathetic tone on the heart thus reducing BP and increasing HRV. The results show that across neuropsychological testing, subjects with WCHT performed poorly in memory tests, whereas subjects with BLH showed slower reaction time, compared with controls.
Recently HRV has also been utilised as a predictor of hypertension . The latter study indicated that control male participants with lower HRV had a greater risk of developing hypertension. Our study showed that subjects with borderline hypertension had lower HRV indices than subjects with WCHTs and normotensives (p > 0.05). It is worth noting that when 24-hour RMSSD was compared against sex, there was significant difference between the three groups (p < 0.05). This may indicate a greater effect of gender on the RMSSD values of subjects than the group type i.e. subject's BP. Therefore it can be predicted that if groups were gender matched, then significant and clinical differences in HRV domains may become apparent. The 24-hour RMSSD correlated significantly with clinic systolic, clinic diastolic, mean night-time diastolic, clinic pulse and mean heart rate values. All these correlations were negative. Thus an increased BP and pulse values was associated with a decreased 24-hour RMSSD (and hence HRV) and; a finding that supports the possible association between high BP and low HRV.
Significant correlation was also evident between four out of the five computer cognition tests. Negative correlation between 24-hour RMSSD, PALT and the SWMT was evident. This suggests that greater number of errors in these tests (i.e. poor cognition) may be linked to low HRV. However, there was positive correlation between 24-hour RMSSD and SSPT and the SRMT. The scores for these tests were based on the number of correct answers, so the positive correlation suggests that low scores (i.e. poor cognition) are linked to low HRV. This correlation entails significant association of low HRV with high BP and poor cognitive performance in this sample, but the differences between the groups were not significant.
Echo cardiography revealed that compared to control subjects, both WCHT and BLH subjects had normal but thicker left ventricular (LV) mass and normal but reduced diastolic function. This is important because detecting signs of organ damage in WCHT may further help physicians to correctly plan antihypertensive therapy, particularly, among patients at higher risk of CVD and other organ damage i.e. CKD from excessive BP reduction .
Taken together, the findings from this study cannot answer the question of whether to treat or not treat WCHT, in part, because of the cross-sectional nature of the study, and indeed the relatively smaller sample size. Nonetheless, the current study may provide new insights on the basis of associated new surrogate risks i.e. cognitive and HRV dysfunction
Firstly, we did not record the exact time of BP measurements in the clinic. Moreover, no standardized measures were taken to reduce the stressor "pressor effects" of the clinic setting. Secondly, unmeasured confounders such as education , may have affected the current cognitive function results. Longitudinal data may be required to produce a clearer understanding of the effects of cognition on clinic and ambulatory BP values over time. Thirdly, the small sample size used in our study may have concealed the true differences in HRV among the groups. The lack of any relationship between group type and HRV was most likely due to the lack of group participants and failure of gender-matching.