Abstract

Background. Congestive heart failure (CHF) is an independent risk factor for mortality in the end‐stage renal disease (ESRD) population. Nocturnal haemodialysis (NHD), a novel mode of renal replacement therapy, may be more effective than conventional haemodialysis in reducing intravascular volume or in removing uraemic toxins with vasoconstrictor or myocardial depressant actions, and may, therefore, improve the left ventricular (LV) systolic function of patients with coexisting cardiac and renal failure.

Methods. To test this hypothesis, we determined, in six patients (mean age±SD: 49.5±9 years), blood pressure (BP), ejection fraction (EF: radionucleotide angiography), left ventricular mass index (LVMI: echocardiography), LV fractional shortening (FS), and extracellular fluid volume (ECFV: bioelectrical impedance): before and after a mean of 3.2±2.1 years following conversion from conventional dialysis (3 days/week×4 h) to NHD (6 nights/week×8–10 h).

Results. There were significant reductions in systolic and mean arterial BP (138±10 to 120±9 mmHg, P=0.04; 99±6 to 86±7 mmHg, P=0.01). There was a significant increase in EF (28±12 to 41±18%, P=0.01) and a trend to greater LV FS (20±10 to 38±17%, P=0.06). Post‐dialysis ECFV was not affected by dialysis mode (18.5±5.1 vs 18.2±3.5 l, P=0.76). The number of prescribed cardiovascular medications was reduced (2.2–0.7, P=0.02).

Conclusions. In ESRD patients with systolic dysfunction, NHD leads to a sustained increase of EF and a reduction in the requirement for vasoactive medications in the absence of any reduction in post‐dialysis ECFV.

Introduction

Cardiovascular events represent the leading cause of death in patients with end‐stage renal disease (ESRD). Congestive heart failure (CHF) is an independent risk factor for premature mortality in this population. The median survival of patients with ESRD and CHF has been estimated to be 36 months, as compared with 62 months for the general dialysis population [1]. CHF may arise from impairment of left ventricular (LV) systolic or diastolic function or both. In the non‐ESRD population, advances in medical therapy have had a major impact on the survival of heart failure patients with systolic dysfunction. In contrast, current medical and renal replacement strategies have not altered mortality rates in those with coexisting cardiac and renal failure [2].

For reasons yet to be established, a single conventional haemodialysis session can improve acutely the LVEF of ESRD patients with depressed systolic function. Possible mechanisms include a decrease in extracellular fluid volume (ECFV), dialysis of myocardial depressant uraemic toxins, and a reduction in atrial pressure, resulting in removal of pericardial constraint on LV filling [35].

Nocturnal haemodialysis (NHD), which provides 8–10 h of renal replacement therapy during sleep, 6–7 nights/week, is a novel and potentially beneficial alternative mode of dialysis for this most vulnerable cohort of patients [6]. Sustained reductions in blood pressure (BP) and regression of LV hypertrophy have been observed following conversion from conventional to NHD [7]. The mechanism or mechanisms responsible for these improvements have not been established definitively, but these may be due to decreases in ECFV and/or total peripheral resistance [8] or perhaps elimination of nocturnal obstructive sleep apnoea [9]. We hypothesized that conversion from conventional haemodialysis to NHD would result in long‐term improvement in LVEF of ESRD patients with coexisting CHF.

Subjects and methods

All patients with impaired ventricular systolic function (EF≤40%) and ESRD, undergoing NHD at the Humber River Regional Hospital since October 1997, were assessed. Patients with clinical evidence of (i) an acute coronary syndrome, (ii) pericardial disease, (iii) myocarditis, and (iv) atrial fibrillation were excluded.

Patients were assessed while on conventional dialysis (baseline) and at 3 monthly intervals after conversion to NHD. Weight and height were recorded. Using the same calibrated sphygmomanometer, BP was measured during clinic visits after 5 min of rest, with patients seated. Serum haemoglobin (Hb) and albumin (Alb) were assessed initially and at 3 monthly intervals.

ECFV was estimated by single‐frequency bioelectrical impedance analysis (BIA; RJL systems device: BIA‐101Q with Fluid and Nutrition Analysis Software 3.2, Clinton, MI) [10]. Electrodes were placed on the wrist and foot for this purpose. Limbs with arteriovenous access were avoided. ECFV was measured 2–3 h after conventional dialysis, and in the morning after a regular session of NHD.

Multiple gated acquisition scans were performed to assess LVEF. LV mass was calculated from 2‐D echocardiographic images according to the formula of Devereux and Reichek [11] at baseline (conventional dialysis) and annually while on NHD. LV mass index (LVMI) was derived by correcting the LV mass for body surface area. LV fractional shortening (FS) was calculated as: FS=(LVED–LVES)/LVED where LVED is LV internal end‐diastolic dimension and LVES is LV internal end‐systolic dimension. Doppler echocardiography was used to grade mitral regurgitation from 1+ to 4+.

Prescribed cardiovascular medications, including diuretics, beta‐blockers, angiotensin converting enzyme inhibitors, angiotensin receptor antagonists, digitalis, calcium channel blockers, and vasodilators, were documented. Following conversion to NHD, withdrawal of vasoactive medications was indicated if patient experienced haemodynamic intolerance (e.g. symptomatic hypotension).

Following conversion, patients received 8–10 h of haemodialysis each night. Vascular access was achieved through either a long‐term internal jugular catheter (Uldall Catheter, Cook Critical Care, Bloomington, IN) or an arteriovenous fistula. A dialysate flow rate of 100–500 ml/min was used. F40, F50, or F80 polysulfone dialyzers (Fresenius Medical Care, Lexington, MA) were used for each treatment.

Primary variables of interest were baseline and the last recorded clinic value for EF, LVMI, FS, and ECF volume. Data are described as mean±SD. Paired Student's t‐test was used for comparison of continuous variables. All statistical tests were two tailed with a P‐value <0.05 taken to indicate significance. SPSS‐10 (Chicago, IL) was used for statistical analyses.

Results

Of the 40 patients presently participating in this programme, eight were identified as having depressed cardiac systolic function. Six patients (mean±SD age 49.5±9 years) met the inclusion criteria (Table 1). Their baseline ejection fraction was 28±12% (range 9–40%) and their ventricular architecture was consistent with eccentric LV hypertrophy (i.e. an increase in LVMI with relatively normal wall thickness). Heart failure in these patients was attributed to ischaemic (n=3), hypertensive (n=2), and dilated (n=1) cardiomyopathy.

Principal variables are summarized in Table 1. The mean duration of NHD was 3.2±2.1 years. After conversion to NHD, there was a fall in systolic (138±10 to 120±9 mmHg, P=0.04) and mean arterial (99±6 to 86±7 mmHg, P=0.01) BP. Diastolic BP tended to decrease (80±9 to 69±7 mmHg, P=0.09). There was a reduction in the number of prescribed cardiovascular medications from 2.2 to 0.7 drugs (P=0.02).

Post‐dialysis ECFV, as measured by BIA was not affected. Alb increased significantly after conversion to NHD whereas Hb tended to increase (Table 1). LVEF increased in all patients (from 28±12 to 42±18%, P=0.01). There was a trend to greater LVFS (from 20±10 to 38±17%, P=0.06). There was a significant reduction in septal wall thickness, and there were trends towards decreased posterior wall thickness and LVMI (from 180±54 to 143±45 g/m2, P=0.1). There was no significant change in the degree of mitral regurgitation.

Table 1. 

Principal variables

Variable
 
Conventional dialysis
 
NHD
 
Systolic BP (mmHg) 138±10 120±9* 
Diastolic BP (mmHg)  80±9  69±7 
Mean arterial pressure (mmHg)  99±6  86±7* 
Pulse pressure (mmHg)  58±16  51±9 
ECFV (l)  18.5±5.2  18.2±3.5 
EF (%)  28±12  42±18* 
LVMI (g/m2180±54 143±45 
LVFS (%)  20±10  38±17 
End‐diastolic diameter (mm)  62±14  59±13 
Septal wall (mm)  10±2   9±1* 
Posterior wall (mm)  10±2   9±0.2 
Mitral regurgitation   1.1±0.8   0.8±0.6 
Hb (g/l) 116±10 131±14† 
Alb (g/l)  35±3  39±3* 
Cardiovascular medications   2.2   0.7* 
Angiotensin converting   5   1 
   enzyme inhibitor   
Angiotensin receptor blocker   1   1 
Beta‐blocker   1   1 
Calcium channel blocker   2   0 
Digoxin   1   1 
Nitrates   1   0 
Diuretic   2   0 
Variable
 
Conventional dialysis
 
NHD
 
Systolic BP (mmHg) 138±10 120±9* 
Diastolic BP (mmHg)  80±9  69±7 
Mean arterial pressure (mmHg)  99±6  86±7* 
Pulse pressure (mmHg)  58±16  51±9 
ECFV (l)  18.5±5.2  18.2±3.5 
EF (%)  28±12  42±18* 
LVMI (g/m2180±54 143±45 
LVFS (%)  20±10  38±17 
End‐diastolic diameter (mm)  62±14  59±13 
Septal wall (mm)  10±2   9±1* 
Posterior wall (mm)  10±2   9±0.2 
Mitral regurgitation   1.1±0.8   0.8±0.6 
Hb (g/l) 116±10 131±14† 
Alb (g/l)  35±3  39±3* 
Cardiovascular medications   2.2   0.7* 
Angiotensin converting   5   1 
   enzyme inhibitor   
Angiotensin receptor blocker   1   1 
Beta‐blocker   1   1 
Calcium channel blocker   2   0 
Digoxin   1   1 
Nitrates   1   0 
Diuretic   2   0 

*P<0.05. †P=0.05. n=6, values represent mean±SD.

Discussion

CHF is an independent risk factor for mortality in the dialysis population [1]. The impact of drug or adjunctive CHF therapies on LVEF or prognosis in the ESRD population has not been widely studied. To our knowledge, this is the first non‐pharmacological intervention to achieve a significant and sustained increase in EF in the haemodialysis population. The magnitude of the increase in EF observed after conversion from conventional haemodialysis to NHD is comparable with that obtained with pharmacological therapies in both dialysis patients with dilated cardiomyopathy [12], and in patients with heart failure due to depressed ventricular systolic function and preserved renal function [13] and those heart failure patients with coexisting central sleep apnoea provided nocturnal continuous positive airway pressure (CPAP) [14].

NHD patients experienced a significant decrease in systolic and mean BP, despite withdrawal of vasoactive drug therapy, and a trend towards lower diastolic BP and LVMI. Because post‐dialysis ECFV was similar before and after conversion to NHD [7], these cardiovascular adaptations are unlikely to arise from an absolute reduction in the nadirs of intravascular blood volume. Consistent with this interpretation is the absence of a significant reduction in pulse pressure.

Although there was no change in post‐dialysis ECFV as determined by BIA, it is reasonable to postulate that the magnitude of daily oscillations in ECFV and the time averaged ECFV decreased after conversion to NHD. These fluctuations may play a key role in the pathophysiology of CHF in ESRD patients and may trigger a number of volume or stretch‐mediated autocrine and paracrine cascades with adverse effects on cardiac systolic function. At the cellular level, cyclic stretch of cardiac fibroblasts and neonatal cardiomyocytes activates growth factors such as transforming growth factor‐beta 1 [15]. In animal models, stimulation of several cardiac and humoral growth factors by volume overload affects systolic function adversely [16]. More detailed physiological studies are required to elucidate the importance of such fluctuations in ECFV, and in particular cardiac chamber volumes, in the progression of CHF in ESRD.

The failing heart is particularly sensitive to changes in afterload. The BP lowering effect of NHD, reported previously [6], was confirmed in the present study. Multiple vasoconstrictor neuro‐endocrine and paracrine systems have already been implicated in the pathogenesis of hypertension in ESRD [8]. These factors may be cleared more effectively by NHD. If so, long‐term reductions in total vascular resistance would be anticipated to improve depressed ventricular systolic function.

There is increasing evidence that accumulation of middle molecules impacts adversely on cardiac performance. The ultrafiltrate and serum of uraemic patients have negative inotropic and chronotropic effects [17]. However, the exact uraemic molecules that exert these cardiotoxic effects have not been determined. In five stable patients with coexisting cardiac and renal failure, Nixon et al. [4] observed an acute increase in EF after a single session of euvolaemic dialysis. These authors concluded that removal of uraemic toxins caused an upward shift of the ventricular function curve (i.e. an increase in contractility without a significant Frank–Starling effect). Parfrey et al. [18] reported normalization of systolic function in 12 ESRD patients with uraemic cardiomyopathy after renal transplantation. NHD offers a significant advantage in uraemia control, and in particular middle molecular clearance [6] when compared with conventional haemodialysis. The present findings are, therefore, consistent with the hypothesis that improvement in uraemia can ameliorate its depressant effect on systolic function.

In the setting of increased right atrial pressure and right ventricular end‐diastolic volume, pericardial constraint and diastolic ventricular interaction assume increasing haemodynamic importance, by inhibiting LV diastolic filling. Atherton et al. [5] studied the effect of acute volume unloading by lower body negative pressure in 12 normal subjects and 21 patients with CHF. LV filling improved in patients with heart failure whereas LVED volume decreased in normal subjects. We hypothesize that NHD might act to decrease right ventricular preload of patients thereby improving LV filling and stroke volume. However, because the post‐haemodialysis ECFV was not altered by this change in dialysis mode, direct haemodynamic measurement would be required to determine whether a long‐term change in diastolic ventricular interaction could contribute to the observed increase in EF.

We did not determine the prevalence of sleep‐related disorders in these six subjects. However, it is known that airway obstruction during sleep triggers acute repetitive increases in LV afterload, decreases in stroke volume, and cycles of hypoxia, hypercapnia, and sympathetic nervous system activation, which, in combination act to accelerate disease progression in heart failure. Elimination of obstructive sleep apnoea by nasal CPAP results in marked improvement in LV systolic function within 1 month of therapy [14]. NHD corrects sleep apnoea associated with ESRD [9]. Further studies are required to elucidate the impact of NHD on afterload and sleep‐related breathing disorders in the ESRD population with CHF.

It is interesting to note that the Hb and Alb levels of our patients improved after conversion to NHD from conventional haemodialysis. The effect of normalization of Hb in chronic haemodialysis patients has been studied previously. LV geometry did not change despite a significant rise in Hb [19]. Therefore, the cardiovascular adaptations are unlikely to arise from an absolute increase in haemoglobin (Hb) levels but rather related to the change in dialysis modality. Malnutrition has been postulated to aggravate pre‐existing heart failure in ESRD patients [20]. In the present study, we observed a significant rise in serum Alb levels in all patients. It is tempting to speculate that improvement in nutritional status as reflected by serum Alb levels contributed to the rise in LVEF. Further studies are required to examine the interplay between malnutrition and cardiac disease in ESRD patients.

To date, there has been no standardized effective management strategy to optimize the care of patients who have co‐existing cardiac and renal failure. In ESRD patients with systolic dysfunction, NHD leads to a sustained increase of EF and a reduction in the requirement for vasoactive medications in the absence of any reduction in post‐dialysis ECFV. A limitation of our study is its observational nature and the limited number of patients. Larger cohort of patients would likely increase the study's power to show a significant difference in parameters such as LVMI and diastolic BP. Further studies are required to elucidate the mechanism of haemodynamic improvement, and the potential impact of NHD on quality of life and survival in this population.

Correspondence and offprint requests to: Dr A. Pierratos, 112 Joicey Blvd, Toronto, Ontario, Canada, M5M 2T6. Email: a.pierratos@utoronto.ca

Dr Christopher Chan holds a Kidney Foundation of Canada Biomedical Fellowship. Dr John Floras holds a Career Investigator Award from the Heart and Stroke Foundation of Ontario. The Nocturnal Hemodialysis Demonstration Project is supported by the Ministry of Health of Ontario, Canada.

References

1
Harnett J, Foley R, Kent G et al. Congestive heart failure in dialysis patients: prevalence, incidence, prognosis and risk factors.
Kidney Int
 
1995
;
47
:
884
–890
2
Canadian Organ Replacement Registry Report: Hospital Medical Records Institute. Don Mills, Ontario, Canada,
2000
3
Hung J, Harris P, Uren R et al. Uremic cardiomyopathy—effect of hemodialysis on left ventricular function in end‐stage renal failure.
N Eng J Med
 
1980
;
302
:
547
–551
4
Nixon J, Mitchell J, McPaul J et al. Effect of hemodialysis on left ventricular function: dissociation of changes in filling volume and in contractile state.
J Clin Invest
 
1983
;
71
:
377
–384
5
Atheron J, Moore T, Lele S et al. Diastolic ventricular interaction in chronic heart failure.
Lancet
 
1997
;
349
:
1720
–1724
6
Pierratos A, Ouwendyk M, Francoeur R et al. Nocturnal hemodialysis: three‐year experience.
J Am Soc Nephrol
 
1998
;
9
:
859
–868
7
Chan CT, Floras JS, Miller JA et al. Regression of left ventricular hypertrophy after conversion to nocturnal hemodalysis.
Kidney Int
 
2002
;
61
:
2235
–2239
8
Luik AJ, Kooman JP, Leunissen KM. Hypertension in hemodialysis patients: is it only hypervolemia?
Nephrol Dial Transplant
 
1997
;
12
:
1557
–1560
9
Hanly P, Pierratos A. Improvement of sleep apnea in patients with chronic renal failure who undergo nocturnal hemodialysis.
N Engl J Med
 
2001
;
344
:
102
–107
10
Charra B, Laurent G, Chazot C et al. Clinical assessment of dry weight.
Nephrol Dial Transplant
 
1996
;
11 [Suppl 2]
:
16
–19
11
Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method.
Circulation
 
1977
;
55
:
613
–618
12
Cice G, Ferrara L, Di Benedetto A et al. Dilated cardiomyopathy in dialysis patients—beneficial effects of carvedilol: a double‐blind, placebo‐controlled trial.
J Am Coll Cardiol
 
2001
;
37
:
407
–411
13
Loh E. Maximizing management of patients with decompensated heart failure.
Clin Cardiol
 
2000
;
3 [Suppl III]
:
1
–5
14
Sin DD, Logan AG, Fitzgerald FS et al. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure patients with and without Cheyne–Stokes respiration.
Circulation
 
2000
;
102
:
61
–66
15
van Wamel AJ, Ruwhof C, van der Valk‐Kokshoom LE et al. The role of angiotensin II, endothelin‐1 and transforming growth factor‐beta as autocrine/paracrine mediators of stretch‐induced cardiomyocyte hypertrophy.
Mol Cell Biochem
 
2001
;
218 (1–2)
:
113
–124
16
Modesti P, Vanni S, Bertolozzi I et al. Early sequence of cardiac adaptations and growth factor formation in pressure—and volume—overload hypertrophy.
Am J Physiol Heart Circ Physiol
 
2000
;
279
:
H976
–985
17
Weissensee D, Low‐Friedrich I, Riehle M et al. In vitro approach to ‘uremic cardiomyopathy’.
Nephron
 
1993
;
65
:
392
–400
18
Parfrey P, Harnett J, Foley R et al. Impact of renal transplantation on uremic cardiomyopathy.
Transplantation
 
1995
;
60
:
908
–914
19
Foley R, Parfrey P, Morgan J et al. Effect of hemoglobin levels in hemodialysis patients with asymptomatic cardiomyopathy.
Kidney Int
 
2000
;
58
:
1325
–1335
20
Bergström J, Lindholm B. Malnutrition, cardiac disease and mortality: an integrated point of view.
Am J Kidney Dis
 
1998
;
32
:
834
–841

Comments

0 Comments