Predictors of Left Ventricular Hypertrophy in Patients with Chronic Kidney Disease

Abstract

The aim of the present study is to determine the prevalence and predictors of left ventricular hypertrophy in patients with stage 3 or 4 chronic kidney disease. Thirty-four patients were included. In addition to hematological and biochemical evaluations, echocardiography and ambulatory blood pressure monitoring were performed both at the beginning and at the end of the first year. Echocardiographic left ventricular mass was calculated and indexed to body surface area to calculate left ventricular mass index (LVMI). Left ventricular hypertrophy was diagnosed if LVMI >131 g/m² in male and >100 g/m² in female patients. During the follow-up period, estimated glomerular filtration rate decreased from 36.6±11.7 to 31.0±14.0 mL/min (p = 0.03), while LVMI increased from 130.2±35.6 to 140.5±30.5 g/m² (p = 0.055). Left ventricular hypertrophy was detected in 67.6% of the patients at the baseline and in 89.7% at the end of the study (p = 0.011). The independent predictors of the final LVMI were age (p = 0.035), baseline day-time systolic blood pressure (p = 0.01), baseline C-reactive protein (p = 0.001), and the decrease in glomerular filtration rate during the follow-up (p = 0.002). Left ventricular hypertrophy is quite frequent among patients with stage 3 or 4 chronic kidney disease, and its prevalence increases while glomerular filtration rate decreases during the follow-up. The early detection of left ventricular hypertrophy and both prevention of the deterioration of renal function and aggressive blood pressure control may help to achieve a decrease in cardiovascular morbidity and mortality in these patients.

Keywords:

• ambulatory blood pressure monitoring
• cardiovascular disease
• chronic kidney disease
• hypertension
• inflammation

INTRODUCTION

Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in patients with end-stage renal disease. The prevalence of coronary artery disease and congestive heart failure is approximately 40% in patients treated by hemodialysis (HD) or peritoneal dialysis (PD), compared with 5–12% in the general population.[1–4] In addition to the traditional risk factors (age, sex, hypertension, diabetes mellitus, smoking), non-traditional risk factors such as anemia, proteinuria, volume overload, hyperparathyroidism, uremic toxins, hyperhomocysteinemia, and malnutrition are shown to be causes of CVD for patients with end-stage renal disease.[2],[3] Both traditional and non-traditional risk factors result in CVD manifested as congestive heart failure, left ventricular hypertrophy (LVH), coronary artery disease, peripheral vascular disease, and myocardial infarction.[4] LVH is an independent predictor of cardiac mortality in dialysis patients as well as in patients with essential hypertension.[5]

Clinical and echocardiographic CVD were found to be highly prevalent at the start of renal replacement therapy.[6] Although the high burden of risk factors that predispose to CVD are well known, the number of studies showing the extent of these risk factors in predialysis patients are limited, as these patients cannot be monitored as closely as the patients treated by dialysis.[7] It is quite important, therefore, to identify the risk factors for the development of CVD during the early stages of chronic kidney disease (CKD). The aim of this study is to search the prevalence and clinical predictors of LVH in patients with CKD.

MATERIALS AND METHODS

The Institutional Review Committee of Ankara University School of Medicine approved the study. Thirty-four patients with stage 3 or 4 CKD who were referred to our outpatient clinics during a three-month period were included. All of the subjects gave informed consent preceding all procedures. During the follow-up, all patients were administrated appropriate treatments to improve their cardiovascular outcome with strict control of blood pressure, correction for anemia and hyperlipidemia, and treatment for hyperparathyroidism. In addition to hematological and biochemical evaluations, office blood pressure measurement, ambulatory blood pressure monitoring (ABPM), and echocardiography were performed both at the beginning and the end of the first year. Age and gender of the patients, etiology, and the duration of the CKD, medications, and body mass index were recorded. Glomerular filtration rate was estimated by MDRD formula. Laboratory evaluations included urinalysis, complete blood count, biochemical parameters (Technikon DAX-96 Chemistry AutoAnalyzer, Bayer Co., USA), serum ferritin (chemiluminescent assay technique, Beckman Coulter, USA), parathyroid hormone (PTH, immuno chemiluminescent assay technique, Immulite 2000 Analyser, DPC, USA), and C-reactive protein (CRP, nephelometric immunoassay, Dade Behring N High Sensitivity CRP Reagent, Dade Behring Inc., USA). Post-active infection treatment CRP values were considered.

ABPM was performed using a non-invasive monitor (Space Lab, Redmond, Washington, USA). Systolic and diastolic blood pressure was recorded every half-hour during the whole day. The mean 24h, daytime, and nighttime systolic and diastolic blood pressure were calculated. Dipping was defined as a reduction in nighttime blood pressure greater than 10% of the daytime blood pressure.

Two-dimension and M-mode echocardiography were performed using Hewlett Packard Sonos 5500 (Hewlett Packard, Andover, Massachusetts, USA) echocardiograph with a 2.5 MHz transducer by the single-experienced echocardiographer, and all of the measurements were made according to the American Society of Echocardiography guidelines.[8] Echocardiographic left ventricular mass was calculated[9] and indexed to body surface area to calculate left ventricular mass index (LVMI). LVH was diagnosed if LVMI >131 g/m² in male and >100 g/m² in female patients.[10]

Statistical analyses were performed using the SPSS software (SPSS for Windows, version 11.5, Chicago, Illinois, USA). Data were analyzed first for normality of distribution using Kolmogorov-Smirnov test. Normally distributed data were expressed as mean ± SD, and non-normally distributed data as median and range. Comparisons of the parameters were done using Student's t-test or Wilcoxon signed rank test, as appropriate. Univariate correlation analyses were done using Pearson's or Spearman's tests, as appropriate. For multiple-factor variables, stepwise multiple regression analysis was used. A two-tailed p less than 0.05 is considered statistically significant.

RESULTS

During the follow-up period, one patient died, and four patients declined to be echocardiographed at the end of the year. Therefore, comparisons were done using the data of the remaining twenty-nine patients. Demographic features and laboratory data of the patients are listed in Table 1. Nineteen patients (55.9%) were on an angiotensin-converting enzyme inhibitor or angiotensin receptor antagonist, 15 patients (44.1%) were on a statin, and eight patients (23.5%) were on aspirin. Fourteen of the patients (41.2%) were smokers. Although none of the patients started dialysis therapy during the follow-up period, glomerular filtration rate significantly decreased from 36.6±11.7 mL/min at the baseline to 31.0±14.0 mL/min at the end of the study (p = 0.03). ABPM data revealed that 24-h diastolic blood pressure significantly decreased during the follow-up (p = 0.017), while 24-h systolic blood pressure did not change significantly (see Table 2). Only four patients were found to be dippers (11.8%).

Table 1 Demographic features of the patients and relevant laboratory data

	Baseline	Final	p
Age (year)	54.4 ± 12.7
Gender (female; n, %)	18 (53%)
Duration of CKD (months)	33.9 ± 15.8
Smoking history (packet-year)	0 (0–45)
Hypertension (n, %)	32 (94.1%)
Duration of hypertension (year)	6.6 ± 6.0
Etiology of CKD (n, %)
Diabetes mellitus	12 (36%)
Hypertension	3 (9%)
Chronic glomerular	7 (20%)
Chronic tubulointerstitial	7 (20%)
Polycystic kidney disease	2 (6%)
Unknown	3 (9%)
Glomerular filtration rate (mL/min)	36.6 ± 11.7	31.0 ± 14.0	0.03
Hemoglobin (g/dL)	12.1 ± 2.1	12.1 ± 2.0	NS
Erythrocyte sedimentation rate (mm/h)	35.0 ± 26.3	28.7 ± 23.7	NS
C-reactive protein (g/dL)	2.8 (0.3–14.3)	3.2 (1–25.0)	NS
Total cholesterol (mg/dL)	198.5 ± 65.6	186.9 ± 55.8	NS
Ferritin (ng/mL)	84 (13–510)	100 (6–1081)	NS
Parathyroid hormone (pg/mL)	160.5 ± 186.2	225.3 ± 183.0	NS
Vitamin B12 (pg/mL)	347.0 ± 185.2	273.3 ± 183.6	NS
Folic acid (ng/mL)	9 (3.3–69.0)	9.2 (6.8–20.0)	NS

Table 2 Ambulatory blood pressure monitoring data and echocardiographic findings

	Baseline	Final	p
24 h systolic blood pressure (mmHg)	130.4 ± 17.0	124.6 ± 14.8	NS
24 h diastolic blood pressure (mmHg)	78.1 ± 9.4	74.3 ± 8.7	0.017
Daytime systolic blood pressure	131.5 ± 17.1	125.4 ± 14.1	NS
Daytime diastolic blood pressure	79.3 ± 10.1	76.0 ± 8.7	NS
Nighttime systolic blood pressure	128.0 ± 18.8	123.6 ± 19.0	NS
Nighttime diastolic blood pressure	74.9 ± 11.1	71.2 ± 11.1	0.034
LVMI (g/m^2)	130.2 ± 35.6	140.5 ± 30.5	0.055
LVH, n (%)	23 (67.6%)	26 (89.7%)	0.011

LVMI increased from 130.2±35.6 g/m² to 140.5±30.5 g/m² (p = 0.055). Twenty-three patients (67.6%) at the baseline and 26 patients at the end of the study (89.7%) had LVH (p = 0.011). Age, erythrocyte sedimentation rate, total cholesterol, and CRP were found to be significantly correlated with LVMI at the baseline. Age (p = 0.006) and erythrocyte sedimentation rate (p = 0.012) were found to be independent predictors of baseline LVMI. The final LVMI was significantly correlated to baseline CRP and total cholesterol, baseline 24-h and daytime systolic blood pressure, office systolic blood pressure both at the baseline and at the end of the study, and the decrease in glomerular filtration rate during the follow-up. There was no significant correlation between dipping status and LVMI. Stepwise multiple linear regression analysis, which includes all the parameters significantly related to final LVMI as the dependent parameter, revealed that age (β, 0.30; 95% confidence interval, 0.053–1.36; p = 0.035), baseline CRP (β, 0.54; 95% confidence interval, 2.20–7.35; p = 0.001), baseline daytime systolic blood pressure (β, 0.40; 95% confidence interval, 2.20–1.37; p = 0.01), and decrease in glomerular filtration rate (β, −0.45; 95% confidence interval, −1.71 – −0.46; p = 0.002) were the independent predictors of the final LVMI.

DISCUSSION

In the present study, echocardiographic findings and ambulatory blood pressure monitoring data from thirty-four patients having stage 3 or 4 CKD were evaluated. At the end of the one-year follow-up period, a significant decrease in glomerular filtration rate was detected together with an increase of marginal statistical level in LVMI. Prevalence of LVH also increased from 67.6% at the baseline to 89.7% at the end of the study. The independent predictors of final LVMI were age, baseline CRP, baseline daytime systolic blood pressure, and decrease in glomerular filtration rate during the follow-up period.

CVD is the major cause of death in patients with end-stage renal disease. The prevalence of clinical and echocardiographic CVD is very high at the start of renal replacement therapy in these patients. High frequency of traditional risk factors and uremia-specific risk factors (known as non-traditional risk factors) for CVD result in congestive heart failure, coronary artery disease, peripheral vascular disease, and LVH.[6] LVH is an independent predictor of mortality in patients with CKD as it has serious clinical consequences, including impaired LV compliance, increased coronary resistance, and arrhythmogenesis.

Clinical and echocardiographic CVD was found in a very high proportion of patients starting renal replacement therapy and were independent mortality factors.[6] Other prospective and retrospective studies also suggest that CVD starts in the very early phases of chronic renal failure. Being consistent with the previous reports, the present study, though having a relatively low number of patients, also revealed a high prevalence of LVH in CKD patients. Moreover, both LVMI and the prevalence of LVH increased during the one-year follow-up period. These findings imply that patients with CKD should be regularly followed by both echocardiography, and the probable risk factors for the development of LVH and treatment strategies against risk factors should be applied from the early stages of the disease. In addition, the finding of the significant correlation between LVMI and the decrease in renal function during follow-up is particularly important and should be confirmed by further studies with a larger number of patients. The strategies against the progression of renal disease therefore may also be of importance in the prevention of LVH.

Hypertension, which is very frequent in CKD patients, is a modifiable risk factor for LVH.[11–14] Abnormal blood pressure diurnal rhythm is also common in uremia, and is shown to be highly correlated with end-organ damage.[15],[16] However, no significant correlations were found between abnormal diurnal rhythm and LVMI. In the present study, and being consistent with the previous reports, hypertension was found as one of the most important predictors of LVMI. In addition, this study also confirms the superiority of ABPM data in the prediction of LVH in predialysis patients. ABPM is not routinely recommended for renal failure patients in hypertension therapy guidelines. On the other hand, when the results of studies with predialysis and dialysis patients are considered, it seems that the application of ABPM in early stages of renal failure patients is mandatory.

No relationship was found between antihypertensive medications and LVH in the present study. Inhibitors of renin-angiotensin-aldosterone system might have specific effects on LVMI, independent of their antihypertensive effect.[17] However, a relatively low number of subjects included in the present study prevent the confirmation of such a result. On the other hand, renin-angiotensin-aldosterone system inhibitors may cause anemia, one of the most prominent determinants of LVH. Indeed, it was previously reported that the withdrawal of ACE inhibitors in HD patients results in an increase in hematocrit level without any significant changes in LVMI.[18]

Inflammation is one of the most prominent risk factors for CVD in the general population and CKD patients.[19–21] In this study, which is consistent with the literature, inflammation markers, erythrocyte sedimentation rate, and CRP were found to be independent predictors of LVMI. Although the mechanism of the relationship between CRP levels and cardiovascular events are still being searched, high proinflammatory cytokines in predialysis patients imply that uremia is a continuous inflammatory condition, and renal failure patients have an increased risk for CVD.[22] Although it is a well-known risk factor for CVD in the general population and chronic renal failure patients, there is no consensus about how to use the CRP levels in renal failure patients.

In conclusion, LVH is quite frequent in patients with stage 3 or 4 CKD, and its prevalence increases while the glomerular filtration rate decreases during the follow-up. Prevention of the deterioration of renal function and aggressive blood pressure control may help to achieve a decrease in cardiovascular mortality and morbidity in these patients. Patients with CKD should be given echocardiography and ABPM regularly from the early stages of the disease.