Cardiovascular Journal of Africa: Vol 32 No 6 (NOVEMBER/DECEMBER 2021)

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 32, No 6, November/December 2021 AFRICA 317 which increases the specificity of identifying prolonged PWD and reproducibility. The prevalence of P max > 120 ms in this study was 31% lower than in that by Asad et al ., 27 who found a prevalence of 47% in a hospitalised population. Methodological differences may account for this. Their patients were an elderly in-patient population with co-morbidities, which alone might have caused increased P max duration and introduced confounders. Our study comprised middle-aged out-patients. In this study, the prevalence of prolonged PWD was 70% higher in the subjects than the 30% found in the controls. This is likely due to the effect of hypertension on the atrial myocardium causing non-uniform contraction, prolonged electrical activation time, left atrial stiffness, increasing left atrial pressure and remodelling. 28-30 Prolonged PWD has also been associated with the development of paroxysmal atrial fibrillation because it reflects an abnormal electrophysiology due to atrial remodelling, stiffness and inhomogeneity in atrial conduction. 29,31 BMI and WC significantly correlated with prolonged PWD and P max in the subjects compared to the controls, which is comparable with the study by Dagil et al . 9 and Nussinovitch et al . 18 They observed that there was a significant difference in obese subjects with a BMI ≥ 30 kg/m 2 . Other studies have shown a significant reduction in PWD with a reduction in BMI after a 12-week weight-loss programme, with a positive correlation between the decrease in level of PWD and percentage weight loss. 32 These observations may suggest that substantial weight loss is associated with improvement in atrial repolarisation, as reflected by improved P-wave indices. In this study, prolonged PWD correlated significantly with increased WC, which is an important independent cardiovascular risk factor. According to Dobbelsteyn et al ., 33 central fat distribution has always been associated with worse cardiovascular risk and WC may be the best single indicator of other multiple cardiovascular risk factors. An association of P max duration and PWD with a sharp increase in blood pressure has been observed. 34 It has also been shown that in hypertensive urgency, the use of medications and subsequent reduction of P max and PWD significantly caused a stability in atrial conduction and haemodynamics. 34 In a subset of pre-hypertensive subjects, 35 it has been shown that PWD increased significantly with rising blood pressure compared to normal blood pressure (< 120/80 mmHg). Mean LVMI was significantly higher among the subjects compared to the controls, so it was also higher among the study population with prolonged PWD than in those with normal PWD. However, comparing within the subject group, between those with prolonged and those with normal PWD, their PWD was not statistically significant. The reason for this is unclear. In this study, there was no significant association between subjects with prolonged PWD and LVMI ( r = 0.08, p = 0.33). This is similar to a Nigerian hypertensive study by Ale et al. where a lack of relationship was found between QT dispersion (QTd) and LVMI. 36 This may be due to the effect of antihypertensive therapy on atrial and ventricular remodelling, 37 as most of the hypertensive patients in this study were already on medication. In a paediatric study on the relationship of PWD, LVMI and blood pressure among school children, 10 it was reported that there was a significant increase in PWD and LVMI among hypertensive children, although some children with a normal LVMI still had increased PWD. This latter finding supports the theory by Martin Gracia et al . 22 that intra-atrial and inter-atrial conduction disorders, expressed by increased PWD in structurally normal hearts, are present before anatomical disturbances appear. In this study, the mean EF was significantly longer in the controls compared to the subjects. There was a significant negative correlation between PWD and EF. With increasing PWD there is a tendency to decreased LV systolic function. This is in agreement with Huseiyin et al ., 14 who reported a weak but significant correlation between PWD and EF. Also Ding et al . 38 reported that PWD improved with an improvement in EF after cardiac resynchronisation therapy. These findings further strengthen the hypothesis that PWD may be used as a non-invasive marker for target-organ damage, as it has been shown in this study that a longer PWD correlated with poorer systolic function. The ability of PWD to reflect diastolic dysfunction may be due to the haemodynamic and structural effects of left ventricular diastolic dysfunction (LVDD) on left atrial structure and function, which is represented by the P wave on ECG. 28 In this study all parameters were considered to determine prevalence of LVDF. The mean E/e ′ , E velocity, E/A ratio and septal e ′ were significantly different among the subjects compared to controls and also different in the proportion of the study population with prolonged PWD. However, comparing the subjects with normal and prolonged PWD, the mean E/e ′ was comparable. Once there is increased LV filling pressure in the setting of hypertension, there is elevated left atrial pressure, which may be associated with inhomogeneous and non-uniform atrial conduction, manifesting as prolonged P-wave duration on ECG and left atrial enlargement. 28 Left atrial enlargement is regarded as a surrogate of LVDD. 39 In this study, mean left atrial diameter was significantly increased in subjects alongside the prolonged PWD. In this study E/e ′ did not correlate clinically with prolonged PWD among the subjects, unlike in the study by Tsai et al ., 28 who reported a significant clinical correlate between corrected PWD and the ratio of transmitral E-wave velocity to early diastolic mitral annulus velocity. Also the prevalence of LVDD was higher in the group with corrected PWD. Although in this study the mean E/e ′ was significantly increased among subjects compared to the controls, there was no statistical difference between the mean E/e ′ of subjects with prolonged and normal PWD. However, Wei-chung et al . 30 found an association of increased arterial stiffness and PWD with LVDD; increased PWD correlated with higher E/e ′ indices, suggesting LVDD. These findings may help us conclude that PWD is a useful screening tool to identify LVDD in high-risk groups such as hypertensives. Conclusion The mean PWD in normal Nigerian subjects was 32.1 ± 14.72 ms and it was significantly lower than in hypertensive patients, at 38.29 ± 8.02 ms. The prevalence of prolonged PWD was 51.3% in the total study population, 70% among subjects and 32.7% among controls. Whereas there was a relationship between PWD and WC, there was no relationship between LVMI and PWD. There was an inverse relationship with PWD and EF, but no association was found between PWD and LVDF using E/e ′ .