Left ventricular mass estimation by different partition values in a large population of black hypertensive subjects

Abstract Aims Our aim is to compare the impact of the 2 most widely used methods of indexing left ventricular mass (LVM) on the distribution of abnormal left ventricular (LV) geometric patterns, in a large sample of untreated asymptomatic black hypertensive subjects. Methods and Results All patients with hypertension referred to the Cardiology unit of University of Abuja Teaching Hospital, Abuja, Nigeria from 2006 to 2013, who gave informed consent, and underwent physical examination and echocardiography. LVM indexation was classified into 4 geometric patterns after echocardiography: normal geometry, concentric hypertrophy, concentric remodeling, and eccentric hypertrophy. Concentric hypertrophy was the commonest geometric pattern and was detected in 33.6% to 39.5% of the patients. LVM/height2.7 was a better method to detect abnormal geometric pattern than LVM/BSA (P < 0.0001). Conclusion In a large cohort of hypertensive subjects with no clinical evidence of cardiovascular disease, abnormal LV geometry was found in greater than four‐fifths of the population. In addition, LVM indexed for height 2.7 was found to be a better method for detecting LVH than LVM indexed for BSA, as the highest prevalence of abnormal geometry was diagnosed when LVM was indexed for height2.7.

2 × PWTd / LVIDD was used to calculate relative wall thickness, and increased RWT was considered present when this ratio exceeded 0.43. 18 For LVM/body surface area (BSA), we used a cut-off value of 116 g/m 2 for men and 104 g/m 2 for women. For LVM/HT 2.7 , we used a cut-off value of 49.2 gm/m 2.7 for men and 46.7 gm/m 2.7 for women. 4 Normal geometry was considered present when there was both normal left ventricular mass index (LVMI) and relative wall thickness, whereas concentric remodeling was identified by normal LVM but increased RWT. Eccentric hypertrophy, on the other hand, was said to be present when there was increased LVMI but normal RWT, while concentric hypertrophy was identified by increased LVMI and RWT.

| Statistical analysis
Data analysis was carried out using SPSS version 16.0 software (SPSS, Chicago IL, USA). Continuous variables are expressed as mean ± SD, while categorical variables are expressed as percentages.
To assess the normality of continuous variables, the Kolmogorov-Smirnov statistics was used. One-way ANOVA with Sheffe's post hoc test was used to assess comparison between the groups, while chisquare was used to compare proportions. A 2-tailed P-value ≤ 0.05 was considered statistically significant.

| RESULTS
3.1 | Clinical and demographic data of the subjects according to the two partition values Table 1 shows that independent of whether LV mass index was estimated using the LVM/BSA or LVM/HT 2.7 indexation method, patients with concentric remodeling were the oldest. Table 2 shows that the largest BMI was found in patients with eccentric hypertrophy, independent of the method of indexation used.
By using LVM/HT 2.7 , we found that patients with eccentric hypertrophy had BMIs that were significantly higher than those in patients with any of the other three geometric patterns. When the LVM/BSA method was used, however, we observed that patients with eccentric hypertrophy had BMIs that were only higher than those in patients with normal geometry. Table 3 shows that female subjects had significantly worse geometric pattern when LVM was indexed for BSA, but there was no difference when indexed for height 2.7 .

| Distribution of LV geometric patterns by percentage in men and women with different modes of LVMI and partition values
Male patients were found to exhibit more concentric hypertrophy than female patients when LVM was indexed for BSA and height 2.7 .
Female patients, on the other hand, exhibited more eccentric hypertrophy than male patients, irrespective of the LVM index method used.
Further, male subjects exhibited more concentric remodeling than female patients when LVM was indexed for height 2.7 compared with when it was indexed for LVM/BSA.

| DISCUSSION
This study in a large cohort of Black hypertensive subjects has revealed a range of abnormal left ventricular geometric pattern from 82% to 90%, depending on the LVM indexation method used. Significantly higher compared with normal geometry (P < 0.001).
b Significantly higher compared with eccentric hypertrophy and normal geometry. c Significantly higher compared with eccentric hypertrophy and normal geometry (P < 0.001).
d Significantly higher compared with concentric hypertrophy, concentric remodeling, and normal geometry (P < 0.001).  Significantly higher compared with normal geometry (P = 0.017).
b Significantly higher compared with eccentric hypertrophy (P < 0.01).
The prevalence of abnormal left ventricular (LV) geometry is higher than that reported in the LIFE study, 19 and by Cuspidi and coworkers. 20  This is also supported by the findings of the Echo Normal study, in which racial differences in the prevalence of LVH were found, in spite of the use of the same cut-off values in defining LVH. 22 Concentric hypertrophy was found to be the most common geometric pattern in this cohort, ranging from 31.8% to 40.6%, similar to the finding of Libhaber and co-workers, 23 who by using the indexation method of LVM indexed for height 2.7 , found concentric hypertrophy to account for approximately 51% of the abnormal geometric pattern in their cohort. We had also earlier demonstrated concentric hypertrophy to be the most common geometric pattern in a small cohort of our hypertensive subjects. 24 The Life study, 19 unlike the present study, showed eccentric hypertrophy as the most common form of abnormal geometry. This can be accounted for by the differences in the 2 populations studied. Apart from the difference in race/ethnicity, subjects in the LIFE study were patients with evidence of target organ damage and presence of clinical cardiovascular disease, while our subjects had no clinically established cardiovascular disease.
Although concentric hypertrophy was the commonest form of abnormal geometry in our cohort, this was based on the 2-tiered classification of LVH, which has some limitations. Therefore, application of the 4-tiered classification of the Dallas Heart study, 25 which sub-classifies concentric and eccentric hypertrophy yielding 4 distinct geometric patterns, as a follow-up to this study, may be a better method of assessing the prevalence of concentric hypertrophy in this cohort.
The highest prevalence of abnormal geometry was diagnosed when LVM was indexed for height 2.7 . The finding that LVM indexed for height 2.7 is a better method for detecting abnormal LV geometry is a very important one, as this indexation method has been reported by De Simone and co-workers to offer the most accurate estimation of LVH and risk factor for pathologic changes in heart structure. 26 Zoccali and co-workers also found LVM indexed for height 2.7 to be a better predictor of cardiovascular events compared with LVM indexed for BSA, in a group of patients undergoing haemodialysis. 27 In addition, LVM to algometric signal of height 2.7 has now been acknowledged by guidelines as the best indexation method in hypertension. 28 Other workers have, however, found that LVM indexed for height 1 [30][31][32] Finally, abnormal geometry was more common in female patients using all 3 indexation methods. This is similar to the findings in the LIFE Study, 14 and those by Adebiyi and co-workers. 10 The most plausible reason for this is the fact that men have larger BSA and are taller than their female counterparts, on average.
Another possible reason for this, as proposed by De Simone et al, is the effect of biological factors especially associated with female fat deposition. 33

| CONCLUSION
In a large cohort of hypertensive subjects with no clinical evidence of cardiovascular disease, abnormal left ventricular geometry was found in greater than four-fifths of the population. In addition, LVM indexed for height 2.7 was found to be a better method than LVM indexed for BSA in diagnosing LVH.