1 Peritoneal Dialysis International, Vol. 20, pp Printed in Canada. All rights reserved /00 $ Copyright 2000 International Society for Peritoneal Dialysis LEFT VENTRICULAR HYPERTROPHY AND CLINICAL OUTCOME IN CAPD PATIENTS Songkwan Silaruks, Dhavee Sirivongs, 1 and Darunee Chunlertrith 1 Divisions of Cardiology and Nephrology, 1 Department of Medicine, Khon Kaen Medical School, Khon Kaen, Thailand Objective: To determine the clinical outcome of left ventricular hypertrophy (LVH) (left ventricular wall diastole thickness 1.2 cm) detected by echocardiography in nondiabetic, continuous ambulatory peritoneal dialysis (CAPD) patients without dilated cardiomyopathy. Design: A prospective, descriptive study was conducted between 1 July 1995 and 31 January Patients were followed up for 24 months. Setting: Peritoneal dialysis unit in a medical school hospital. Patients and Methods: Baseline and yearly echocardiograms were carried out on 66 patients receiving CAPD. Cardiac death was assessed. LVH was correlated with outcome. Results: Of 66 nondiabetic CAPD patients without dilated cardiomyopathy, 20 had a normal echocardiogram (LV wall thickness < 1.2 cm), 21 had mild hypertrophy, and 25 severe hypertrophy (LV wall thickness > 1.4 cm in diastole). In the first two groups, 21% were admitted with congestive heart failure (CHF) after starting dialysis. The 1-year cumulative survival was 85% among those with mild hypertrophy and 91% in the normal group. In the group with severe hypertrophy, 57% were admitted at least once with CHF, and the 1-year cumulative survival was 56%. Eighty-two percent of those who died in the severe group, which accounted for the significantly worse survival (p = 0.003), died from cardiac or cerebrovascular causes, compared with none of those with a normal echocardiogram. Conclusions: Severe LVH was found in a third of our CAPD patients and was associated with a significantly high cardiovascular morbidity and mortality. KEY WORDS: Left ventricular hypertrophy; echocardiography; cardiac mortality; outcomes. Cardiovascular disease remains a leading cause of morbidity and mortality in patients with endstage renal disease (ESRD), accounting for 40% of Correspondence to: D. Sirivongs, Department of Nephrology, Khon Kaen Medical School, 123 Friendship Road, Khon Kaen Thailand. Received 6 July 1999; accepted 28 March 2000 deaths in international registries (1,2). Ischemic heart disease, cerebrovascular disease, and left ventricular hypertrophy (LVH) all occur frequently in the dialysis population (2). In the Framingham Study, LVH, although infrequently noted in electrocardiograms (ECG), has long been recognized as an independent predictor of subsequent cardiovascular morbidity and mortality (3 7). The advent of echocardiography has established not only an accurate noninvasive method for detection of LVH, but preliminary studies have also implicated echocardiographic LVH as a potent predictor of cardiovascular morbid events and stroke (8 11). Congestive heart failure (CHF), in addition to age and diabetes mellitus, is an adverse prognostic indicator in ESRD patients (12), and is associated with hypertrophic hyperkinetic cardiomyopathy, as well as dilated cardiomyopathy and ischemic heart disease (13). As the natural history of LVH detected by echocardiography in Thai dialysis patients has not been studied, we prospectively studied 66 nondiabetic, chronic peritoneal dialysis patients without dilated cardiomyopathy at the Khon Kean University Medical School Hospital (Srinagarind). This was followed by two echocardiograms carried out annually to determine the clinical and echocardiographic outcomes of LVH. PATIENTS AND METHODS The patient population comprised all nondiabetic continuous ambulatory peritoneal dialysis (CAPD) patients (n = 68) treated at Srinagarind Hospital. Diabetic patients (n = 18) were excluded from the analysis because cardiovascular mortality is higher among diabetic than nondiabetic patients on dialysis (13). Two nondiabetic CAPD patients who had dilated cardiomyopathy were excluded from the analysis. Patients who did not attend for follow-up or who had a successful transplant were excluded from the study. Subject entered the study after the first echocardiogram was carried out. The study was conducted between 1 July 1995 and 31 January
2 SILARUKS et al. JULY 2000 VOL. 20, NO. 4 PDI ECHOCARDIOGRAPHIC METHODS Echocardiography was performed by the first author using M-mode and two-dimensional sonography, using the criteria set forth by the American Society of Echocardiography (14), at entry and at 12-month intervals. The echocardiographic measurements included left ventricular (LV) end-diastolic diameter, LV end-systolic diameter, left atrial diameter, LV wall thickness in diastole, and LV ejection fraction. These values were used in the diagnosis (see definitions below). After the study had begun, annual serial echocardiograms were performed on all patients. The data obtained from the hospital records were age, gender, duration of treatment with peritoneal dialysis, current ESRD treatment, original disease, and current medication. The following investigations were undertaken: blood hemoglobin, serum albumin (bromcresol green method) and fasting cholesterol, chest x-rays, and ECGs. Twenty-four months later, the records were again reviewed to obtain information on admissions to hospital or death. The following definitions were used prior to data analysis: 1. Congestive heart failure: a history of dyspnea and peripheral edema, with cardiomegaly and two of the five following signs: raised jugular venous pressure, basal crepitations, peripheral edema, pulmonary venous hypertension, or interstitial edema on a chest x ray (12). 2. Ischemic heart disease: a history of angina and/or myocardial infarction, the latter confirmed by ECG changes or an increase in cardiac enzymes. Coronary artery disease was considered to be present if a myocardial infarction had occurred, bypass surgery had been undertaken, or the disease had been demonstrated on angiography or at autopsy. 3. Normal echocardiogram: LV end-diastolic diameter < 5.5 cm, ejection fraction > 55%, and LV wall thickness in diastole < 1.2 cm. 4. Dilated cardiomyopathy: LV-end diastolic diameter 5.5 cm, and ejection fraction < 50% (15). 5. Mild hypertrophy: ejection fraction > 55%, LV wall thickness 1.2 cm and < 1.4 cm in diastole (16). Left ventricular hypertrophy is commonly defined as LV wall thickness 1.2 cm in diastole (17). 6. Severe hypertrophy: ejection fraction > 55%, LV wall thickness 1.4 cm in diastole (16). 7. Hypertrophic hyperkinetic disease: LV end-diastolic diameter < 5 cm, ejection fraction > 70%, and LV wall thickness in diastole 1.4 cm (12). 8. Uncontrolled hypertension: systolic blood pressure > 180 mmhg or diastolic blood pressure > 110 mmhg, while taking 3 antihypertensive agents Cardiac mortality: cause of death recorded as myocardial infarction, sudden death, or other cardiac causes. STATISTICS For normally distributed continuous variables, the mean values and standard deviations were calculated and the means compared using the two-sample t-test. Categorical variables were compared using the chisquare test or Fisher s exact test. To determine variables that were most significantly and independently related to outcome, a multiple logistical regression analysis was carried out. The life table survival was obtained by using the Kaplan Meier method, while the log-rank model was used for statistical analysis. All statistical tests were two-tailed, with a p value less than 0.05 taken to indicate statistical significance. Logistic regression and survival analyses were undertaken using Biomedical Data Processing Software (18) and applying both Mantel Cox and Breslow statistical tests. RESULTS Sixty-six nondiabetic CAPD patients without dilated cardiomyopathy were recruited into the study. Thirty percent (20 patients) had no LVH at the time of entry, 32% (21 patients) had mild concentric hypertrophy, and 38% (25 patients) had severe hypertrophy. Of the latter, 48% (12 patients) had hypertrophic hyperkinetic disease. None of our patients had a functioning vascular access or a prior history of hemodialysis. Table 1 presents the clinical variables of the population studied. Patients were between 26 and 79 years old, and 66% were male. The predominant causes of renal failure were glomerulonephritis (37%), hypertension (29%), and tubulointerstitial disease (19%). Reflux nephropathy, pyelonephritis, and polycystic kidney disease were the other predominant causes. Echocardiographic dimensions for the three study groups are shown in Table 2. In the group with normal echocardiograms, 14 patients (70%) had an echocardiogram performed at the start of, or within the first year of, dialysis (median duration of dialysis at time of echocardiogram was 11 months). In the group with mild hypertrophy, 12 patients (60%) had the test performed during the same period (median duration of dialysis was 12 months). Of those with severe hypertrophy, 20 patients (80%) had the test done during the same period (median duration of dialysis was 10 months), p = 0.45.
3 PDI JULY 2000 VOL. 20, NO. 4 LVH IN CAPD PATIENTS TABLE 1 Demographic and Clinical Variables of Peritoneal Dialysis Patients With and Without Left Ventricular Hypertrophy (LVH) Normal Mild LVH Severe LVH (mean ±SD) (mean ±SD) (mean ±SD) (n=20) (n=21) (n=25) p Value a Age (years) 44.6± ± ± Male/female 17/3 15/6 17/8 N S Duration of ESRD therapy (years) 4.2± ± ±3.9 N S Systolic BP (mmhg) 140±16 142±21 146±23 N S Diastolic BP (mmhg) 72±11 79±14 81±13 N S Hemoglobin (g/dl) 83±2 87±9 84±7 N S Cholesterol (mg/dl) 203.1± ± ±35.3 N S N (%) N (%) N (%) Age > 50 years 4 (2.0) 7 (33.3) 12 (8.0) 0.04 Male population 17 (85.0) 15 (71.4) 17 (68.0) N S Cardiomegaly 7 (35.0) 10 (47.6) 14 (56.0) N S LVH by ECG 0 (0) 1 (4.7) 2 (8.0) N S Hypoalbuminemia (<3.5 mg/dl) 15 (75.0) 14 (66.7) 19 (76.0) N S Hospitalization for CHF 0 (0) 2 (9.5) 13 (52.0) 0.01 Acute myocardial infarction 1 (5.0) 3 (14.3) 1 (4.0) N S Uncontrolled hypertension 1 (5.0) 2 (9.5) 2 (8.0) N S ESRD = end-stage renal disease; CHF = congestive heart failure; NS = not significant. a Normal versus severe LVH. TABLE 2 Echocardiographic Results in Patients with No LVH (Normal), Mild, and Severe LVH at Start of Study (Mean ± SD) Dimensions (cm) Normal Mild LVH Severe LVH Left atrium 3.4± ± ±0.4 Left ventricular end-diastolic 4.4± ± ±0.6 Left ventricular end-systolic 2.9± ± ±0.6 Interventricular septum in diastole 0.9± ± ±0.3 Left ventricular wall in diastole 0.9± ± ±0.3 ECHOCARDIOGRAPHIC OUTCOME (TABLE 3) In the group whose initial echocardiogram was normal, the only significant change in dimensions at the end of study was a diastole cm thickening of the LV wall, after 19 ± 5 months (mean ± SD) of follow-up. Only 1 patient in this group subsequently developed severe hypertrophy. In the group with mild hypertrophy on the initial echocardiogram, 2 patients (10%) progressed from mild to severe hypertrophy during the 24 months of follow-up. There were no significant differences in the clinical, demographic, or biochemical variables between those that progressed to severe disease and those that had persistent mild hypertrophy. Of the 25 patients with severe LVH, there was no significant variation in mean values of cardiac dimensions, and the vast majority of the severe group continued to have severe hypertrophy at the end of the study. CLINICAL OUTCOME After starting ESRD therapy, a significantly higher proportion of patients with severe hypertrophy was admitted to hospital at least once with CHF (Table 1). None of the patients had a renal transplant during the study. Mortality: The effect of LVH on cumulative patient survival over 2 years in this population is presented in Figure 1. The survival rate of patients with severe LVH was significantly less than that of patients with a normal echocardiogram (p = 0.003) and of patients with only mild LVH (p = 0.01). The 1-year cumulative survival rates in those with normal, mild, and 463
4 SILARUKS et al. JULY 2000 VOL. 20, NO. 4 PDI TABLE 3 Echocardiographic Outcome in Peritoneal Dialysis Patients Final echocardiogram Normal Mild LVH Severe LVH D C M Total Initial echocardiogram a N (%) N (%) N (%) N (%) N (%) Normal 15 (75) 4 (20.0) 1 (5.0) 0 (0) 20 (30.3) Mild LVH 0 (0) 17 (80.9) 2 (9.5) 2 (9.5) 21 (31.8) Severe LVH 0 (0) 0 (0) 23 (92.0) 2 (8.0) 25 (37.9) Total 15 (22.7) 21 (31.8) 26 (39.4) 4 (6.1) 66 (100) LVH = left ventricular hypertrophy; DCM = dilated cardiomyopathy. a The initial echocardiogram was that performed at the start of the prospective study. Figure 1 Percent cumulative survival in peritoneal dialysis patients is shown. Survival of patients with severe left ventricular hypertrophy (LVH) was significantly worse than those with normal echocardiogram (p = 0.003) and than those with mild LVH (p = 0.01). severe LVH were 91%, 85%, and 56%, respectively. Differences in survival rates persisted even when differences in age were taken into account. The 1-year cumulative survival of patients aged less than 50 who had a normal echocardiogram (n = 8) was 100%, and in those with severe hypertrophy (n = 14), 72% (p = 0.03). In patients aged 50 years or older who had a normal echocardiogram (n = 12), the 1-year survival was 83%, and in those with severe hypertrophy (n = 11), 36% (p = 0.02). Of the patients who had mild LVH at the start of the study and subsequently progressed to either severe LVH or dilated cardiomyopathy, 25% (1 of 4) died, compared with 12% who did not progress. This did not reach statistical significance. The baseline characteristics of those who did and did not progress to severe LVH are presented in Table 3. There were no significant differences in clinical variables between the two groups. The causes of death in the three groups are shown in Table 4. Eighty-two percent of the severe LVH group died from either cardiac or cerebrovascular disease, compared with none of the group with a normal echocardiogram (p = 0.01). The numbers of deaths from causes other than cardiac or cerebrovascular disease were similar in the groups with normal echocardiography (2 deaths), mild LVH (2), and severe LVH (2). DISCUSSION There is a scarcity of data on the cardiovascular risk for patients entering a long-term CAPD program. It has been shown that there is a similar mortality rate between cardiovascular CAPD and hemodialysis patients (19). This is somewhat surprising considering the theoretical advantages of this technique relevant to hemodynamic stability. These advantages include the absence of an arteriovenous fistula, which, together with renal anemia, is one of the main causes of hypercirculation. In contrast to electrocardiographic LVH, which occurs in only 3.2% of the general population (3), LVH determined by echocardiography was a common finding in the Framingham study, where it was docu- TABLE 4 Causes of Death in Peritoneal Dialysis Patients Cardiac Cerebrovascular Other Mortality Echocardiogram (N) N (%) N (%) N (%) (%) Normal (20) 0 (0) 0 (0) 2 (10.0) 10.0 Mild LVH (21) 1 (4.8) 0 (0) 2 (9.5) 14.3 Severe LVH (25) 6 (24.0) 3 (12.0) 2 (8.0)
5 PDI JULY 2000 VOL. 20, NO. 4 LVH IN CAPD PATIENTS mented in 16% of men and 19% of women (20), in 17% 42% of patients with hypertension (21), and in 68% of dialysis patients (16). The Framingham study also showed that during 4 years of follow-up, echocardiographic LVH was a predictor of further coronary events and stroke (9), a conclusion corroborated by Aronow et al. (10). In our study, severe LVH was not only associated with cardiac morbidity but also with decreased survival (Figure 1). The adverse prognosis of severe LVH is demonstrated by the difference in mortality between those with a normal echocardiogram and those with severe LVH who died of cardiac and cerebrovascular diseases. The observed difference in mortality cannot be explained by the differences in other prognostic factors, other than LVH itself. The most important adverse prognostic indicators of dialysis patients are age, diabetes mellitus, and CHF (12). Diabetic patients were not included in our study. Differences in mortality between the two groups persisted when differences in age were taken into account. Congestive heart failure in dialysis patients is associated with hypertrophic hyperkinetic cardiomyopathy, dilated cardiomyopathy, and ischemic heart disease (13). We excluded those with dilated cardiomyopathy, so the differences in mortality cannot be ascribed solely to ischemic heart disease. Only 12% (3 of 25) of those with severe LVH had a previous myocardial infarction. Only 2 of 9 patients with severe LVH, who had selective coronary angiography during hospitalization, had significant coronary artery disease. However, the small number of patients is insufficient to analyze statistically the relationship between the underlying cardiomyopathy, coronary artery disease, and the difference in mortality. Our study started with a group of patients who were already on dialysis. In patients with a normal echocardiogram or mild LVH on initial echocardiogram, only 7.3% progressed to severe LVH. Harnett et al. (16) and Eisenberg et al. (22) identified hypertension as an important predictor of severe LVH, but could not find any clinical variables that could entirely explain why some dialysis patients develop LVH and others do not. In the current study we were unable to identify any risk factor (such as hypertension, hypoalbuminemia, or anemia) that predisposed a patient to severe LVH while on peritoneal dialysis. We have chosen to define LVH by left ventricular wall thickness, rather than left ventricular mass, because these definitions are frequently used in clinical practice and by other researchers (16,17). Our study is limited by a variable starting point, small patient numbers, and inadequate data on other possible factors that can influence survival (coronary angiographic study, autopsy results). We did not make a systematic collection of the indices of dialysis adequacy which reflects the treatment in use when the study began. Despite its limitations, our study is likely to be unique, providing some information for studying a peritoneal dialysis population with respect to LVH, whereas most previous LVH studies have concentrated on hemodialysis patients. CONCLUSION Severe LVH frequently occurs in patients on peritoneal dialysis. The condition persists, predisposes to the development of CHF, and is associated with high mortality. Its pathogenesis is not clear, and whether severe LVH has a direct causal effect on survival requires further study. ACKNOWLEDGMENTS The authors gratefully acknowledge the assistance of Professor Guy Edwards and Mr. Bryan Hamman in improving the English and the presentation of the data. REFERENCES 1. United States Renal Data System. USRDS 1991 annual report. Bethesda, MD: The National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Fassbinder W, Brunner FP, Brynger H, Ehrich JH, Geerlings W, Raine AE, et al. Combined report on regular dialysis and transplantation in Europe, XX, Nephrol Dial Transplant 1991; 6(Suppl 1): Kannel WB, Gordon T, Offutt D. Left ventricular hypertrophy by electrocardiogram. Prevalence, incidence and mortality in the Framingham Study. Ann Intern Med 1969; 71: Kannel WB, Doyle JT, McNamara PM, Quickenton P, Gordon T. Precursors of sudden coronary death. Factors related to the incidence of sudden death. Circulation 1975; 51: Kannel WB, Dawber TR, Kagan A, Revotskie N, Stokes J III. Factors of risk in the development of coronary heart disease six year follow-up experience: the Framingham Study. Ann Intern Med 1961; 55: Kannel WB, Gordon T, Castelli WP, Margolis JR. Electrocardiographic left ventricular hypertrophy and risk of coronary heart disease: the Framingham Study. Ann Intern Med 1970; 72: Kannel WB. Prevalence and natural history of electrocardiographic left ventricular hypertrophy. Am J Med 1983; 75(3A): Casale PN, Devereux RB, Milner M, Zullo G, Harshfield GA, Pickering TG, et al. Value of echocardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann Intern Med 1986; 105: Levy D, Savage DD, Garrison RJ, Anderson KM, Kannel WB, Castelli WP. Echocardiographic criteria for left 465
6 SILARUKS et al. JULY 2000 VOL. 20, NO. 4 PDI ventricular hypertrophy: the Framingham Heart Study. Am J Cardiol 1987; 59: Aronow WS, Koenigsberg M, Schwartz KS. Usefulness of echocardiographic left ventricular hypertrophy in predicting new coronary events and atherothrombotic brain infarction in patients over 62 years of age. Am J Cardiol 1988; 61: Hutchinson TA, Thomas DC, MacGibbon B. Predicting survival in adults with end-stage renal disease: an age equivalence index. Ann Intern Med 1982; 96: Parfrey PS, Harnett JD, Griffiths SM, Gault MH, Barre PE. Congestive heart failure in dialysis patients. Arch Intern Med 1988; 148: Ritz E, Strumpf C, Katz F, Wing AJ, Quellhorst E. Hypertension and cardiovascular risk factors in hemodialyzed diabetic patients. Hypertension 1985; 7(Suppl 2): II Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurement. Circulation 1978; 58: Parfrey PS, Harnett JD, Griffiths S, Gault MH, Barre PE, Gultmann RD. Low-output left ventricular failure in end-stage renal disease. Am J Nephrol 1987; 7: Harnett JD, Parfrey PS, Griffiths SM, Gault MH, Barre P, Guttmann RD. Left ventricular hypertrophy in endstage renal disease. Nephron 1988; 48: Carr AA, Prisant LM, Watkins LO. Detection of hypertensive left ventricular hypertrophy. Hypertension 1985; 7: Dixon WJ, Brown MB, Engelman L, Hill MA, Henrich RI, eds. BMPD statistical software computer programs. Health Sciences Computing Facility. Berkeley: University of California Press, Gokal R, Jakubowski C, King J, Hunt L, Bogle S, Baillod R, et al. Outcome in patients on continuous ambulatory peritoneal dialysis and haemodialysis. A 4-year analysis of a prospective multicentre study. Lancet 1987; 2: Levy D, Anderson KM, Savage DD, Kannel WB, Christiansen JC, Castelli WP. Echocardiographically detected left ventricular hypertrophy: prevalence and risk factors. The Framingham Heart Study. Ann Intern Med 1988; 108: Devereux RB, Pickering TG, Alderman MH, Chien S, Borer JS, Laragh JH. Left ventricular hypertrophy in hypertension. Prevalence and relationship to pathophysiologic variables. Hypertension 1987; 9(Suppl 2): II Eisenberg M, Prichard S, Barre P, Patton R, Hutchinson T, Sniderman A. Left ventricular hypertrophy in endstage renal disease on peritoneal dialysis. Am J Cardiol 1987; 60: