Changes in left ventricular hypertrophy and function in hypertensive patients started on continuous ambulatory peritoneal dialysis

Changes in left ventricular hypertrophy and function in hypertensive patients started on continuous ambulatory peritoneal dialysis Continuous ambulatory peritoneal dialysis (CAPD) often leads to better control of hypertension. In order to evaluate the effects of such improved blood pressure control on left ventricular (LV) hypertrophy and LV function, a group of 18 patients with a history of hypertension were followed for changes in LV anatomy and function (with M-mode echocardiography) over a 8 to 12 month period after initiation of CAPD. All patients had echocardiographic evidence of increased LV mass related to concentric and eccentric hypertrophy. On CAPD, blood pressure decreased (X mm Hg) in 12 patients. LV mass decreased in 15 patients and increased in one. A decrease in both wall thickness and LV dimension contributed to the fall in LV mass on CAPD. Initially, LV dimension exceeded normal in g out of 18 patients. On CAPD, LV dimension decreased to near normal in size in six, and no patient developed LV dilation on CAPD. Four patients initially had a decreased fractional shortening and ejection fraction; three of these normalized while on CAPD and no patient deteriorated. These results indicate that CAPD improves LV hypertrophy by normalizing both volume and pressure overload. These effects may prevent deterioration in LV function in patients with still normal LV function, and may improve LV function in patients who already exhibit decreased LV performance. (AM HEART J 110:102, 1985.)

Frans H. H. Leenen, M.D., Donna L. Smith, R.N., Ramesh Dimitrios G. Oreopoulos, M.D., Toronto, Ontario, Canada

Several studies1e5 have shown that a large percentage of patients with end-stage renal disease have left ventricular hypertrophy (LVH) and LV dysfunction. Prolonged LVH leads to depressed contractile performance, abnormal compliance, and ultimately LV failure.6 In addition, LVH potentiates the clinical manifestations of coronary artery disease, particularly the incidence of sudden death.7 On the other hand, several studies have shown that normalization of blood pressure by antihypertensive pharmacotherapy can cause a regression of LVH.8-” It seems likely that regression of LVH by adequate antihypertensive treatment will reduce the incidence of hypertension-related cardiac disease, or From the Hypertension Division of Nephrology, pital and Department

Unit, Division of Clinical Pharmacology Department of Medicine, Toronto Western of Medicine, University of Toronto.

This study was supported from the Ministry of Health Leenen) from the Canadian Received accepted

for publication March 4, 1985.

Reprint requests: Frans of Clinical Pharmacology, Toronto, Ontario M5T

102

by Provincial Health Research of Ontario, and by a Research Heart Foundation. Aug.

30, 1984;

revision

H. H. Leenen, M.D., Toronto Western 2S8, Canada.

Grant PR 890 Scholarship (Dr.

received

Hypertension Hospital,

and Hos-

399

Feb.

1, 1985;

Unit, Division Bathurst St.,

Khanna,

M.D., and

more specifically, will prevent deterioration of LV function and improve decreased LV function. ambulatory peritoneal dialysis Continuous (CAPD) often leads to better control of hypertension, allowing a reduction or discontinuation of antihypertensive drug therapy.12 If CAPD results in more optimal blood pressure control and is not associated with volume overload6 and/or increased sympathetic drive,13 one may expect that existing LVH should regress and depressed LV function improve. To test this hypothesis, we followed a group of patients for changes in LV anatomy and function after initiation of CAPD. METHODS LV echocardiograms were obtained in all consenting patients in the Toronto Western Hospital dialysis unit who had end-stage renal disease and a history of hypertension; patients were only admitted to the study if LV echocardiograms of adequate technical quality could be obtained. Eighteen patients qualified; their mean age was 44 years (range 21 to 72). A history of manifest coronary artery disease was present in six, and four patients had diabetes mellitus. Patients entered the study either while on intermittent

Volume Number

110 1, part

Regression

1

Mean

Blood (mm Hg)

130

-

110

-

so

-

70

-

LVH

on ambulatory

peritoneal

dialysis

103

Pressure

I

I

1 1

wk

I 1

IPD Fig.

of

I

mo

6

Ino*

12

I nlQ6

CAP0

1. Changesin supine mean arterial blood pressureon CAPD for individual patients (n = 18).

peritoneal dialysis (IPD; duration varying from 2 weeksto 10 months), or within the first week following initiation of CAPD. Follow-up echocardiogramswere obtained at 3 to 4 weeksand at 6 and 12 months on CAPD. For a number of reasons,not all follow-up studies could be obtained in all patients. Echocardiography. Echocardiograms were obtained with the subjects in the supine position using a Picker Echoview System 60 C, with a 2.25MHz, 13mm diameter, medium focus transducer in conjunction with a Honeywell strip-chart recorder (Fiber optics-cathode ray tube visicorder oscillographModel L5-6A). Tracings wererecorded at 50 mm/set paper speed and measurementswere made to the nearest millimeter on at least four cardiac cycles during quiet respiration; the mean results were used for analysis. All echocardiogramsin a subject were obtained by the same technician with the patient in the same position at the sametime of the day in the sameintercostal area and in the sameLV areas,just below the tip of the mitral leaflets. Measurements were made by the same observer in accordancewith the guidelines of the American Society of Echocardiography.14This observer was blinded regarding previous measurements and the study phase of the patient. The following parameters were measuredor calculated: LV end-diastolic and end-systolic dimension, septal and LV posterior wall thickness (systolic and diastolic), fractional shortening, and ejection fraction. LV masswas determined from intracavitary LV end-diastolic volume and total LV volume (including myocardial volume) according to: LV mass= 1.05 x (total - intra-cavitary volume), where the volumes were obtained by the cube method.15 Data analysis. Results shown on the graphs are for the

individual patients. Because of the number of missing data, comparisonswere done between the first and last measurementsonly, using paired t tests. Statistical analysis was, moreover, done by chi square test for nominal variables and by stepwiseregressionanalysis.‘” RESULTS Blood pressure and LV mass. At initiation of CAPD, 10 patients were still hypertensive (blood pressure [BP] >140/90 mm Hg; Fig. 1). On CAPD, BP of these patients decreased rather consistently (first vs last measurement: -14 ?z 2.9 mm Hg, p < 0.01). One of the normotensive patients showed the opposite: a change from normotension (81 mm Hg) to mild hypertension (113 mm Hg). At the first study 10 out of 18 patients were on antihypertensive drug therapy @-blocker +- vasodilator). Such drug therapy could be discontinued in four patients within a few

weeks on CAPD, therapy. LV mass (Fig. patients, in some decreased in 15

and none required

an increase in

2) was initially increased in all markedly. On CAPD, LV mass out of 18 patients (first vs last measurement of all patients, p < 0.05). In two patients no changes were noted. In one patient (on minoxidil therapy) LV mass actually increased despite adequate blood pressure control. Including the value for this patient, LV mass showed a decrease of 18 + 3.7% or 93 + 32.4 gm (p < 0.02, values at 6 to 12 months vs initial value); without this patient the decrease was 120 gm.

July,

104

Leenen et al. LV (9’

American

!,” End-llia~t hd

Heart

1985

Journal

Run

Fig.

2. Changesin LV massand in LV wall thickness (septum + posterior wall) on CAPD for individual patients (n = 18). N = upper limit of normal.

Fig. 3. Changesin LV end-diastolic and systolic dimension on CAPD for individual patients (n = 18). N = upper limit of normal.

The fall in LV mass was due to decreases in both LV wall thickness (Fig. 2) and LV end-diastolic dimension (Fig. 3). LV wall thickness (septum + posterior wall) was initially increased in all patients; on CAPD it decreased in 14 of 18 patients (p < 0.02). Two patients showed no change, and two patients showed an increase (in one of these, LV mass remained unchanged, and in the other LV mass clearly increased). Including the value of the latter patient, LV wall thickness showed a decrease of 1.7 t 0.84 mm (p < 0.05) or 5.5%; without this patient the decrease was 2.3 + 0.65 mm Cp < 0.01). Septal and posterior wall thickness contributed equally to the initial LVII (16 + 0.5 and 16 & 0.4 mm, respectively) and decreased on CAPD similarly (-1 + 0.4 and -1 st_0.3 mm, respectively; both p < 0.05). LV end-diastolic dimension (LVEDD) decreased in 13 out of 18 patients (p < 0.04) and showed small (<3 mm) increases in five. Including all patients, LVEDD decreased by 5 ? 1.5 mm or 10% (p < 0.01).

A stepwise regression analysis approach was used to describe the relationship between changes in LV mass on CAPD and changes in mean BP and in LVEDD millimeters and milliliters) and initial LVEDD (millimeters and milliliters). Together these variables had a multiple correlation coefficient r = +0.87 (p < 0.01) with the changes in LV mass. Following removal of the mean BP, the multiple correlation coefficient remained at r = +0.85. Stepwise removal of variables indicated that only initial LVEDD (in milliliters) and the change in LVEDD (in milliliters) were significant variables in the regression (individual r = +0.72 and +0.70, respectively) . Presence or absence of antihypertensive therapy while on CAPD, coronary artery disease, or of diabetes mellitus did not affect the extent of changes in LV mass on CAPD. LV function. As shown in Fig. 3, LVEDD exceeded the upper limit of normal in 9 out of 18 patients on IPD or at the start of CAPD. While on CAPD, six out of these nine dilated hearts decreased to near

Volume

110

Number

1, part

1

normal in size (p = 0.16, ns). None of the patients who initially had a normal-sized heart showed development of LV dilation on CAPD; in fact, most of these hearts decreased further in size. LV endsystolic dimension decreased in 12 out of 18 patients 0, = 0.07); five patients showed no change or small (<4 mm) increases (Fig. 3). The fractional shortening of the LV during systole was initially normal in 14 out of 18 patients, and remained in the normal range in all while on CAPD (Fig. 4). Four patients initially had a decreased fractional shortening. Three of these normalized while on CAPD; only one did not show any appreciable improvement. Ejection fraction changed in a fashion parallel to that of fractional shortening. Four patients initially had a decreased ejection fraction; three of these normalized while on CAPD and one remained below normal (Fig. 4).

Regression

of LVH

on ambulatory

peritoneal

dialysis

105

I r

DISCUSSION

In the present series of patients, we confirm the previous observation l2 that use of CAPD to treat patients with end-stage renal disease results in a more optimal contro! of BP. Although this finding is clinically very important, it is equally of interest to demonstrate that improved BP control also leads to beneficial changes in target organs. For example, studies in hypertensive rats have clearly demonstrated that normalization of BP by arterial vasodilators does not improve LVH17; if anything, LVH increases, possibly due to the hyperdynamic circulation. In the present study only one patient had a further rise on LV mass while on CAPD, and this occurred despite adequate BP control; this patient was taking the arterial vasodilator, minoxidil. LV mass on CAPD. BP control in hypertensive humans has generally been shown to lead to regression in LVH.‘-” The present results indicate that this also occurs in hypertensive patients with endstage renal disease when they are treated with CAPD. Most patients exhibited a decrease in LV mass while on CAPD,, with some of the changes being rather marked and resulting in close to normal values for LV mass. Increased LV mass was related to both concentric (increased wall thickness) and eccentric (increased end-diastolic dimension) hypertrophy. Concentric hypertrophy is caused by an increase in afterload, because of arterial hypertension. Eccentric hypertrophy is caused by volume overload, presumably because of anemia and sodium and water retention. The fall in LV mass which occurred on CAPD was due to an improvement in wall thickness, but especially due to a decrease in LV dimension. The stepwise regression analysis

I 1

IPO

rrk

1

mo

6

mo.

12

I mo.

CAPCJ

Fig. 4. Changes in LV fractional shortening and ejection

fraction on CAPD for individual patients (n = 18). N = lower limit of normal.

indicates that the fall in LV mass is only to a small extent related to more optimal BP control and is mainly related to decreases in LV end-diastolic dimensions. In other studies (Leenen et al., submitted) we found that plasma and blood volume, as well as cardiac output, decrease significantly on CAPD as compared with IPD. This decrease in volume will diminish preload and LV internal dimension. Improved LV emptying related to the fall in BP will also result in a smaller LV. The stimulus for eccentric hypertrophy of the cardiac muscle is therefore decreased by CAPD. These results may also indicate that volume overload was a major factor in the etiology of the LVH (possibly by increasing both pre- and afterload) in these patients with end-stage renal disease. Control of the volume overload by CAPD may decrease LVH by decreasing volume overload (LVEDD) and-via a fall in cardiac output-by a fall in afterload (BP). LV function on CAPD. Two parameters of LV function (ejection fraction and fractional shortening) normalized in three of’ four patients with initially

July,

106

Leenen et al.

decreased function, and no patient showed a deterioration while on CAPD. Some of the patients had coronary artery disease and hence may have had dyskinetic areas; therefore, the absolute values obtained with M-mode echocardiography may be incorrect in these patients. However, unless new dyskinetic areas developed during the study (and there was no clinical or echocardiographic evidence of this), relative changes should still be valid. The improvement in LV function observed on CAPD can at least in part be related to the fall in afterload, which enables the LV to empty more easily. The decrease in LVH may have been a contributing factor by permitting an improvement in the actual contractile performance of the LV. Finally, less use of antihypertensive agents with negative inotropic effects may also have improved decreased LV function. Conclusions. The present study shows that initiation of CAPD can lead to marked improvement in LVH and LV function of hypertensive patients with end-stage renal disease within a few months. These effects of CAPD appear to be related to its beneficial effects on BP and especially on volume overload of the heart. REFERENCES

1. Vaziri ND, Prakash R: Echocardiographic evaluation of the effect of hemodialysis on cardiac size and function in patients with end-stage renal disease. Am J Med Sci 278:201, 1979. 2. Friedman HS, Shah BN, Kim HG, Bove LA, DelMonte MM, Smith AJ: Clinical study of the cardiac findings in patients on chronic maintenance hemodialysis: The relationship to coronary risk factors. Clin Nephrol 16:75, 1981. 3. Ikaheimo M, Huttonen K, Takkunen J: Cardiac effects of chronic renal failure and haemodialysis treatment. Hypertensive versus normotensive patients. Br Heart J 45:710, 1981. 4. MacDonald IL, Uldall R, Buda AJ: The effect of hemodialysis

American

Heart

on _--_cardiac rhythm and performance. Clin Nephrol

1995

Journal

15:321,

1Ytll.

5. Montague TJ, MacDonald RPR, Boutilier FE, MacLeod AJ, Cohen AD, Smith ER: Cardiac function in end-stage renal disease. Chest 82441, 1982. 6. Grossman W: Cardiac hypertrophy: Useful adaption or pathologic process? Am J Med 69:576, 1980. 7. Koyanagi S, Eastham C, Marcus ML: Effects of chronic hypertension and left ventricular hypertrophy on the incidence of sudden cardiac death after coronary artery occlusion in conscious dogs. Circulation 65:1192, 1982. 8. Schlant RC, Felner JM, Heymsfield SB, Gilbert CA, Shulman NB, Tuttle EP, Blumenstein BA: Echocardiographic studies of left ventricular anatomy and function in essential hypertension. Cardiovasc Med 2~477, 1977. 9. Dunn FG, Bastian B, Laurie TDV, Lorimer AR: Effect of blood pressure control on left ventricular hypertrophy in patients with essential hypertension. Clin Sci 5934419, 1980. 10. Khatri I, Gottdiener J, Notargiacomo A, Freis E: Effect of therapy on left ventricular function in hypertension. Clin Sci 5934358,198O.

11. Fouad FM, Nakashima Y, Taraxi RC. Salcedo EE: Reversal of left ventricular hypertrophy in hypertensive patients treated with methyldopa. Lack of association with blood pressure control. Am J Cardiol 49:795, 1982. 12. Khanna R, Oreopoulos DG, Dombros N, Vas S, Williams P, Meema HE, Husdan H, Ogilvie R, Zellerman G, Roncari WAK, Clayton S, Izaff S: Continuous ambulatory peritoneal dialysis (CAPD) after three years: Still a promising treatment. Peritoneal Dialysis Bull 1:24, 1981. 13. Ostman-Smith I: Cardiac sympathetic nerves as the final common pathway in the induction of adaptive cardiac hypertrophy. Clin Sci 61:265, 1981. 14. Sahn DJ, DeMaria A, Kisslo J, Weyman A: Recommendations regarding quantitation in M-mode echocardiographic measurements. Circulation 58:1072, 1978. 15. Devereux RB, Reichek N: Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method. Circulation 55:613, 1977. 16. Kleinbaum DG, Kupper LL: Applied regression analysis and other multivariable methods. Boston, 1978, Duxbury Press, p 227.

17. Sen S, Tarazi RC, Bumpus FM: Cardiac hypertrophy and antihypertensive therapy. Cardiovasc Res 11:427, 1977.