Introduction
The prevalence of cardiovascular disease (CVD) is high in the hemodialysis (HD) population [1,2]. CVD is the leading cause of morbidity and mortality in patients on dialysis, contributing to more than 50% of deaths and accounting for approximately 10 to 20 times higher risk in these patients than in the general population [3,4]. Left ventricular hypertrophy (LVH) is a crucial independent predicting factor of adverse cardiovascular outcomes, including myocardial infarction, sudden death, stroke, congestive heart failure, and CVD mortality [5,6]. The prevalence of LVH increases with the reduction of kidney function: with 16%–32% in individuals for estimated glomerular filtration rate (eGFR) >30 mL/min, which elevates to 60%–75% at dialysis initiation and reaches approximately 61%–86.7% in patients on maintenance dialysis [7–11].
LVH is an adaptive remodeling progressive condition that responds to an increase in cardiac work, which might be associated with increased afterload, increased preload, or both. Increased peripheral resistance, elevated blood pressure, increased arterial stiffness, and reduction of large-vessel compliance cause left ventricular (LV) wall thickening and subsequent concentric LV remodeling. However, volume overload causes LV chamber enlargement and subsequent eccentric LV remodeling. Because of the unsteady status of fluid volume and cyclic variation of electrolyte balance in patients on dialysis, it is difficult to distinguish eccentric from concentric patterns in this population [11,12].
LVH in chronic kidney disease (CKD) is also associated with renal anemia and CKD-related mineral and bone disorder (MBD). Anemia in CKD results from relative erythropoietin deficiency [13], leading to increased cardiac output, potentially causing LVH and worsening cardiac damage [11]. Patients with CKD-MBD often exhibit hypocalcemia, hyperphosphatemia, vitamin D3 deficiency, elevated parathyroid hormone (PTH), and fibroblast growth factor 23 (FGF23). Notably, vitamin D3 deficiency and increased FGF23 are related to LVH development [14–18].
According to the American Society of Echocardiography [19], LVH severity can be determined using left ventricular mass index (LVMI), and the strong relationship between increasing LVMI and a higher risk of cardiovascular events has been confirmed by several studies [20,21]. The MAVI (Massa Ventricolare sinistra nell’Ipertensione arteriosa) study revealed that each 39 g/m2 increase in left ventricular mass (LVM) causes a 40% increase in the risk of cardiovascular events [22]. By contrast, in a cohort study of 153 patients on HD receiving parallel treatment for hypertension and anemia, LVM attenuation was independently related to a decrease in all-cause and cardiovascular mortality [23].
A previous study investigated the type of LVH with sudden cardiac death among patients on chronic HD [10], but it did not explore the effect of LVH geometric patterns and severity on all-cause mortality among these patients. Therefore, in the present study, we investigated the prevalence and geometric distribution of LVH in Asian patients on HD and explored the correlation of LVH geometric patterns and severity with cardiovascular mortality, all-cause mortality, and major adverse cardiac events (MACEs).
Methods
The study protocol was approved by the Institutional Review Board of Shin Kong Wu Ho-Su Memorial Hospital (No. 20220713R), and the requirement for informed consent was waived.
Study participants
This retrospective cohort study was conducted at the HD unit of Shin Kong Wu Ho-Su Memorial Hospital in Taipei. We selected 299 adults (aged ≥18 years) who underwent maintenance HD and received echocardiographic examination between October 1 and December 31, 2018. Maintenance HD was defined as undergoing HD for >3 months. Based on LVMI, patients with or without LVH were divided. Then, we stratified patients by the LVH geometric pattern (concentric or eccentric) and subdivided them into four groups based on a combination of LVH geometric patterns and severity (mild-to-moderate eccentric LVH, mild-to-moderate concentric LVH, severe eccentric LVH, and severe concentric LVH). Fig. 1 illustrates the patient selection process.
Echocardiographic measurement
An echocardiographic examination was performed by an experienced cardiologist through two-dimesional echocardiographic linear measurements. Measurements were obtained according to the recommendations of the American Society of Echocardiography and the European Association of Cardiovascular Imaging [24]. The definition of LVH was sex-related, with LVMI >95 g/m2 in women and LVMI >115 g/m2 in men. LVM was calculated using the Cube formula: LVM = 0.80 × [1.04 (IVS + LVID + PWT)3 − (LVID)3] + 0.6, and LVMI = LVM/body surface area. Interventricular septum (IVS), left ventricular internal dimension (LVID), and posterior wall thickness (PWT) were measured at the end of diastole. LVH geometric patterns were defined on the basis of relative wall thickness (RWT), with RWT = (2 × PWT)/(LVID). Concentric LVH was defined as RWT > 0.42, and eccentric LVH was defined as RWT ≤ 0.42. LVH was further classified into mild, moderate, and severe based on LVMI with the recommended sex-specific cutoff values: mild (96–108 g/m2 for women, 116–131 g/m2 for men); moderate (109–121 g/m2 for women, 132–148 g/m2 for men); and severe (>121 g/m2 for women, >148 g/m2 for men) [19]. Left ventricular ejection fraction (LVEF) was calculated using echocardiography software.
Study characteristics and outcomes
Demographics including age, sex, weight, dialysis vintage, and laboratory values including hemoglobin (Hb), serum uric acid, ionic calcium, serum phosphate, total cholesterol, triglyceride, intact PTH, ferritin, and Kt/v (Gotch) were collected after patient enrollment. History of diabetes mellitus (DM), hypertension, dyslipidemia, CVD, chronic heart failure, cerebrovascular accident, and peripheral arterial disease (PAD) and medication history, including renin-angiotensin system inhibitor (RASI), beta-blocker, calcium-channel blockers, vasodilator, alpha-blocker, statin, oral antidiabetic agents, and insulin analog, were all obtained from medical reports.
This study evaluated the association of LVH geometric patterns and severity with clinical outcomes, including cardiovascular mortality, all-cause mortality, and MACEs. A MACE was considered as the composite of death from cardiovascular causes, and hospitalization due to nonfatal myocardial infarction, nonfatal stroke, and coronary revascularization. Enrolled patients were followed up until the occurrence of clinical outcomes or the end of our study, which was defined as December 31, 2021.
Statistical analysis
Continuous variables are summarized as mean ± standard deviation, and categorical variables are expressed as number and percentage. The Kolmogorov-Smirnov test was performed to test the normal distribution for each variable. Comparisons among the two and four groups were performed using the Student t test (normally distributed) or the Mann-Whitney U test (non-normally distributed), one-way analysis of variance (normally distributed), or Kruskal-Wallis test (non-normally distributed) for continuous variables, respectively. The Pearson chi-square test (or Fisher exact test for values less than 5) was used to compare categorical variables. Univariate and multivariable Cox proportional hazard analyses were performed to estimate the crude hazard ratios (cHRs) and adjusted hazard ratios (aHRs) of cardiovascular mortality, all-cause mortality, and MACE associated with LVH type as well as LVH severity and LVH type, respectively.
The cumulative incidence of cardiovascular mortality, all-cause mortality, and MACE during the follow-up period for eccentric and concentric LVH or groups divided by LVH type and LVH severity were evaluated using the Kaplan-Meier method and were compared using the log-rank test. A p-value of <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 22 for Windows (IBM Corp.).
Results
Participant clinicodemographic characteristics
We initially selected 299 patients on chronic HD who completed the echocardiography and clinical evaluation, and LVH was present in 237 patients (79.3%; mean age, 69.04 ± 12.18 years; 52.3% men and 47.7% women). Details regarding the clinical, demographical, and echocardiographic data for all investigated patients were provided in Supplementary Table 1 (available online). Of the remaining 237 patients with LVH, 131 had concentric hypertrophy, and 106 had eccentric hypertrophy. A total of 195 patients (82.3%) had hypertension, and 116 (48.9%) had DM. The median dialysis vintage was 5 years, and patients were followed up for a median of 3.25 years (range, 0.16–3.25 years). Table 1 presents the clinicodemographic parameters at baseline, with patients stratified by concentric or eccentric hypertrophy. Female preponderance was observed in the concentric LVH group compared with the eccentric LVH group (53.4% vs. 40.6%, p = 0.049). Compared with patients with eccentric hypertrophy, those with concentric hypertrophy had significantly longer dialysis vintage (p = 0.04) and a higher prevalence of PAD (33.6%, p = 0.002). Significant differences between groups were also observed in Hb levels (p = 0.02).
Of the 237 patients with LVH, 33 had mild-to-moderate eccentric LVH, 44 had mild-to-moderate concentric LVH, 73 had severe eccentric LVH, and 87 had severe concentric LVH. Table 2 presents their clinicodemographic characteristics. Compared with other groups, female preponderance was noted in the severe concentric LVH group (p = 0.03). A higher percentage of patients with PAD and lower Hb levels were found in the concentric group, irrespective of mild-to-moderate or severe grade (p = 0.009 and p = 0.04, respectively). In addition, patients with severe concentric LVH were more likely to use RASI (71.3%, p = 0.02) compared with those with severe eccentric LVH (60.3%), mild-to-moderate eccentric LVH (48.5%), or mild-to-moderate concentric LVH (45.5%).
Table 3 presents the echocardiographic measurements for patients with LVH, stratified by LVH geometric patterns and severity. LVM, LVMI, PWT, RWT, and LVID were all significantly different between the groups (p < 0.001), and the LVEF was lower in severe LVH, regardless of concentric or eccentric patterns (p = 0.02).
Association of left ventricular hypertrophy with cardiovascular mortality, all-cause mortality, and major adverse cardiac events
Supplementary Fig. 1 (available online) shows the Kaplan-Meier survival analysis of the participants grouped by LVH. Supplementary Fig. 1A (available online) revealed that LVH was significantly associated with MACEs (log-rank p = 0.03). Compared with patients without LVH, those with LVH have higher risks of CV mortality and all-cause mortality, but the result was not statistically significant (log-rank p = 0.120 and log-rank p = 0.441, respectively). Supplementary Table 2 (available online) showed the Cox proportional hazard analysis of outcome events between groups according to LVH presence.
Association of left ventricular hypertrophy type or left ventricular hypertrophy type and severity with cardiovascular mortality
During the 3-year follow-up period, 53 deaths were due to cardiovascular events, including deaths in 37 patients with concentric LVH and 16 with eccentric LVH. Compared with the eccentric LVH group, the HR of cardiovascular mortality was 2.03 (95% CI, 1.13–3.65) in the concentric LVH group. After adjustments for age and sex or for age, sex, vintage, PAD, and Hb level, the risk of cardiovascular mortality remained significant for concentric hypertrophy (aHRs, 2.35 [95% CI, 1.29−4.25] and 2.13 [95% CI, 1.13–4.01], respectively) (Table 4). During the follow-up period, Kaplan-Meier survival analysis revealed that cardiovascular mortality significantly increased in patients with concentric hypertrophy than in those with eccentric geometry (log-rank p = 0.02) (Fig. 2A). Among the 53 patients with cardiovascular mortality, 3, 11, 13, and 26 had mild-to-moderate eccentric LVH, mild-to-moderate concentric LVH, severe eccentric LVH, and severe concentric LVH, respectively. Patients with severe concentric LVH had the highest risk of cardiovascular mortality (HR, 3.73; 95% CI, 1.13–12.33), followed by those with mild-to-moderate concentric LVH (HR, 3.03; 95% CI, 0.85–10.85), severe eccentric LVH (HR, 2.06; 95% CI, 0.59–7.23), and mild-to-moderate eccentric LVH. These differences remained significant after adjustment for age and sex or for age, sex, PAD, Hb, and the percentage of patients using RASI use (Table 4). The Kaplan-Meier curves for the cumulative incidence of cardiovascular mortality in patients with different grades of concentric and eccentric LVH are presented in Fig. 2D, which indicates that cardiovascular mortality was higher in the severe concentric LVH group, but there was no statistically significant difference (log-rank p = 0.07).
Association of left ventricular hypertrophy type or left ventricular hypertrophy type and severity with all-cause mortality
Overall, 65 deaths were recorded, with 42 deaths in the concentric LVH group (12 and 30 in mild-to-moderate and severe concentric LVH groups, respectively) and 23 in the eccentric LVH group (5 and 18 in the mild-to-moderate and severe eccentric LVH groups, respectively). Compared with patients with eccentric LVH, the HR of all-cause mortality was 1.6 (95% CI, 0.96–2.66) in those with concentric LVH. The results are presented in Table 4. After adjustments for age and sex or for age, sex, vintage, PAD, and Hb level, the aHR of all-cause mortality was 1.83 (95% CI, 1.09–3.06) and 1.68 (95% CI, 0.96–2.93), respectively. The Kaplan-Meier curves for the cumulative incidence of all-cause mortality in patients with concentric and eccentric LVH are presented in Fig. 2B, which revealed that all-cause mortality was slightly higher in the concentric group, but there was no statistically significant difference (p = 0.07). The HR of the all-cause mortality of patients with severe concentric LVH was higher (2.58; 95% CI, 1.00–6.65; p = 0.05) (Table 4). After adjustment for age and sex, the difference remained significant (p = 0.02), and the trend remained similar after adjustments for age, sex, PAD, Hb, and the percentage of patients using RASI. The Kaplan-Meier curves for the cumulative incidence of all-cause mortality in patients with different grades of concentric and eccentric LVH are presented in Fig. 2E, and severe eccentric LVH had the worst outcome (log-rank p = 0.18).
Association of left ventricular hypertrophy type or left ventricular hypertrophy type and severity with major adverse cardiac events
Overall, 96 cases of MACEs were observed; of them, 54 cases occurred in the concentric LVH group (18 and 36 in mild-to-moderate and severe concentric LVH groups, respectively) and 42 occurred in the eccentric LVH group (eight and 34 in mild-to-moderate and severe eccentric LVH groups, respectively). Compared with patients with eccentric LVH, the HR of MACEs was 1.07 (95% CI, 0.96–2.66) in those with concentric LVH (Table 4). After adjustments for age and for sex or for age, sex, vintage, PAD, and Hb level, the aHR of MACEs was 1.18 (95% CI, 0.78–1.77) and 1.01 (95% CI, 0.65–1.57), respectively. The Kaplan-Meier curves for the cumulative incidence of MACEs in patients with concentric and eccentric LVH are presented in Fig. 2C (log-rank p = 0.73). According to Table 4, no significant result for MACEs was observed in the four subgroups, even after adjustment for age and sex or for age, sex, PAD, Hb, and the percentage of patients using RASI (all p ≥ 0.05). Fig. 2F presents the results of the Kaplan-Meier analysis for the MACEs in patients with different grades of concentric and eccentric LVH (log-rank p = 0.33).
Discussion
Our findings revealed a high prevalence of LVH in patients on chronic HD, showed that patients with LVH exhibit a higher incidence of MACEs, indicated that concentric hypertrophy was the predominant geometry, and that the concentric LVH group had increased risks of cardiovascular and all-cause mortality. Moreover, patients with severe concentric LVH were at the highest risk of cardiovascular mortality, followed by mild-to-moderate concentric LVH, severe eccentric LVH, and mild-to-moderate eccentric LVH; a similar trend was observed for all-cause mortality.
Previous studies already proved that LVH elevated the incidence of MACEs, and this finding was confirmed in our investigation [25]. However, the results of all-cause mortality and CV mortality were not statistically significant in our study, which might be affected by the small number of patients without LVH.
Many patients with concentric LVH in our study had hypertension, DM, or both. Studies have demonstrated that concentric hypertrophy is the most common geometric pattern in the hypertensive group and DM population. The Resist-POL study, comprising 155 patients with resistant hypertension, revealed a high prevalence of concentric LVH (33%), followed by concentric remodeling (25.8%), normal geometry (24.4%), and eccentric hypertrophy (16.8%) [25]. Another study also found that the prevalence of concentric hypertrophy was higher than that of eccentric patterns in untreated hypertensive patients [26]. Nardi et al. [27] confirmed that in patients with hypertension and CKD, concomitant DM is associated with increased LV wall thickness and higher concentric geometry. Both hypertension and DM are vital factors contributing to concentric LVH. Eguchi et al. [28] noted stronger associations between higher wall thickness and higher prevalence of concentric LVH in patients with type 2 DM than in those without DM. The development of LVH in DM occurs through multiple potential mechanisms, including hyperglycemia-associated cellular alterations, oxidative stress, inflammation, insulin resistance, AMP-activated kinase, and mechanistic target of rapamycin signaling [29]. In particular, cardiac steatosis, decreased myocardial energetics, and systolic dysfunction are potential mechanisms underlying concentric LV remodeling in the DM population [30].
Consistent with our findings, a study conducted in China involving 131 chronic HD patients also observed that concentric LVH (n = 71) was more prevalent than eccentric LVH (n = 9) [8]. By contrast, the CONvective TRAnsport Study (CONTRAST) reported that concentric patterns were mainly observed in patients with CKD not undergoing HD, whereas eccentric hypertrophy was predominant in patients on HD [10,31]. The inconsistency between our investigation and the CONTRAST may be related to the high percentage of patients with hypertension (82.3%) and DM (48.9%) in our study compared with only 26.8% of patients with LVH diagnosed as having DM in the previous study [31]. Given that our finding is consistent with those of the Chinese study, but not with those of the Western study, the influence of ethnicity cannot be neglected. Future large-scale studies are warranted to determine the potential explanations underlying these findings.
Notably, compared with eccentric LVH, concentric hypertrophy has an adverse prognosis in terms of cardiovascular mortality and all-cause mortality. Koren et al. [32] investigated patients with essential hypertension and reported that those with concentric hypertrophy had the highest risks of cardiovascular events and death, whereas those with normal geometry had the lowest risks. In a prospective study, Muiesan et al. [33] found that the persistence or development of concentric LVH was the strongest predictor of cardiovascular events in the hypertensive population. The close relationship between concentric geometry and cardiovascular events might be explained by impaired myocardial contractility, severe diastolic filling abnormalities, increased oxygen consumption, a higher risk of arrhythmias, and sudden death. In addition, abnormal activation of the renin-angiotensin-aldosterone system leads to pathological muscular fibrosis and excessive vasoconstriction, which is associated with subsequent cardiovascular complications [34]. Mulè et al. [35] proved that elevated plasma aldosterone levels may contribute to the development of a concentric LVH in hypertensive patients with CKD.
Only a few studies have investigated the association between LV geometry and adverse events in the HD population. In the CONTRAST, a higher risk of sudden death (adjusted HR, 5.22; 95% CI, 1.14–23.94; p = 0.03) was noted in the eccentric LVH group than in the concentric LVH group; however, the incidence of all-cause mortality or cardiovascular death (adjusted HRs, 0.87; p = 0.57 and adjusted HRs, 1.49; p = 0.57, respectively) was not significantly different between the two groups [10]. The inconsistent results might be attributed to several reasons. First, de Roij van Zuijdewijn et al. [10] only distinguished patients by geometric patterns and not LVH severity, the inconsistencies in the LVH severity division between different geometries might cause nonsignificant results for all-cause mortality and cardiovascular death. Second, we could not determine the patients’ fluid status; stable volume control may explain the lower risks of cardiovascular mortality and all-cause mortality in patients with eccentric patterns. Furthermore, the shorter dialysis vintage in the eccentric LVH group may also influence the prognosis.
In patients with end-stage kidney disease, LVH progresses with dialysis vintage [36]. Zoccali et al. [37] enrolled 161 patients on regular dialysis without a history of congestive heart failure for 18 months and showed that LVMI increased by 7% at the end of the study. Oguz et al. [38] followed up 80 patients on HD and found that LVMI was positively correlated with dialysis vintage (r = 0.387, p = 0.005). Only 32.5% of patients had mild-to-moderate (concentric or eccentric) LVH in our study, which might be related to longer dialysis vintage. Moreover, several studies have demonstrated a strong relationship between increasing LVMI and poor clinical outcomes. In a study of 40,138 adults by 'Bouzas-Mosquera et al. [20], patients with severe increase in LVM had the highest 10-year mortality (46.4%), followed by those with moderately increased LVM (37.4%), mild increase in LVM (31.9%), and normal LVM (26.8%, p < 0.001). However, the risk of clinical endpoints was decreased by the reduction in LVMI. A prospective cohort substudy including 941 patients aged 55 to 80 years with essential hypertension and LVH from the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial had LVM measured by echocardiography. At the 4.8-year follow-up, lower in-treatment LVMI decreased cardiovascular mortality by 38% and all-cause mortality by 28% [39].
To the best of our knowledge, ours is the first study exploring the association of LVH geometric patterns and severity with clinical outcomes, including cardiovascular mortality, all-cause mortality, and MACEs in patients on chronic HD. As mentioned earlier in the text, increasing LVMI was strongly related to a higher risk of cardiovascular events, and concentric patterns had worse outcomes than eccentric patterns. This may explain why the severe concentric LVH group had the poorest prognosis among the four groups.
This study has some limitations. First, this study was a single-center, retrospective, and nonrandomized study, which may have inherent shortcomings such as selection bias and unaccounted confounders. Not every patient undergoing HD received echocardiography, only patients who were available and willing would receive the examinations. Second, our sample size was small, and uneven distribution of some clinical parameters at baseline could have potentially played confounding or collinearity effects. Our study results have the potential for overfitting, coefficient instability, and biased estimates due to some Cox regression analyses having less than 10 events per variable which required caution in interpreting these results and the magnitude of effects might be exaggerated. Third, although all the examinations were performed by professional cardiologists, echocardiography is a highly operator-dependent technique, and we could not exclude operating differences. Fourth, the follow-up time of echocardiography varies, which might be pre- or post-dialysis, and may affect the fluid status in our patients. Fifth, other factors that may affect the outcomes, such as the severity of comorbidities, quality of life, and patient compliance, were not evaluated because of incomplete information. Finally, we didn’t investigate whether these patients had some systemic diseases, such as amyloidosis and Fabry disease, which may also lead to LVH and influence the outcomes.
Our results revealed that concentric LVH was predominant in patients on chronic HD, and that the LVH geometric pattern affected prognosis: patients with concentric LVH had a worse risk of cardiovascular mortality than those with eccentric LVH. Moreover, we demonstrated that LVH progressed with longer dialysis vintage. Our analysis provides novel information on clinical outcomes associated with hypertrophy geometry and LVH severity in patients on HD. Our results suggested that severe concentric LVH was related to the highest risks of cardiovascular and all-cause mortality, followed by mild-to-moderate concentric LVH, severe eccentric LVH, and mild-to-moderate eccentric LVH. However, further large-scale studies are needed to evaluate the role of LVH geometry in HD patients and also to evaluate the potential influence of ethnicity in determining different forms of LVH.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement table 1
Print
