End stage renal
disease (ESRD), the final stage of chronic kidney disease (CKD) is the failure
or near failure of the kidneys to perform their normal functions. 1 These
functions include: excretion, maintenance of acid-base, fluid and electrolyte
balance and the synthesis of hormones such as erythropoietin and renin.     When a patient reaches ESRD,
renal replacement therapy (RRT) in the form of dialysis or transplantation must
be considered, as without treatment, symptoms will likely significantly
deteriorate within weeks or months. There are two major forms of dialysis that
are available, i.e., HD
(Hemodialysis) and peritoneal dialysis (PD). They differ in terms of technique
and physiology.      HD requires vascular access while PD
requires peritoneal access. HD
is an intermittent form of RRT that results in removal of toxins, electrolytes,
and fluid in a periodic pattern, while CAPD is a continuous process. The choice
of modality depends on individual choice, patient’s abilities,
medical problems, or geographic location.      According to
estimates, less than 20% of all long-term dialysis patients are on PD. The two
factors that were perceived to militate against PD were cost and high infection
rates.     Cardiovascular
disease (CVD) is the leading cause of mortality among the patients on dialysis. Dialysis may cause changes in
hemodynamics, electrolytes, oxygenation, acid-base balance and autonomic
balance all of which are potentially arrhythmogenic. 2-5*      USRDS-matched data shows that HD
patients during the first 06 months of therapy are at increased risk of SCD,
may be because patients who are susceptible to adverse events on HD have these
events early. After this initial period, the event rates for HD and peritoneal
dialysis (PD) merge up to 2 years, after which PD patients are more at risk
(Rich et al 2006). 9*     The early stages
account for most of the individuals with CKD but because individuals with any
stage of CKD have a higher risk of cardiovascular disease morbidity and
mortality than their risk of progression to ESRD, cardiovascular risk factor
management in this group is critical.           Unfortunately, from
India there is limited data available on the prevalence of CKD. Indian population-based study determined the
crude and age-adjusted ESRD incidence rates at 151 and 232 per million
population respectively. 2 Diabetes mellitus and hypertension constitute over 2/3rd of
CKD cases in western countries. 11 In India
too, diabetes and hypertension today account for as much as 40–60% cases of
CKD. 12     Standard hemodialysis is
performed three times a week, for 3–5 hours each session, with low-flux
membranes. Since the 1960s there has been a general technical improvement in
dialysis machines, dialyzers, and dialysis solutions. These factors, as well as
accurate control of the ultrafiltration rate by new dialysis machines and more
precise control of electrolyte dialysate composition, have improved hemodynamic
stability during treatment. The substitution of acetate as the dialysate buffer
with bicarbonate has also improved vascular stability. The principle of hemodialysis
is relatively straightforward. Blood flows on one side of a semi permeable
membrane, and dialysis fluid, an osmotically balanced solution of electrolytes,
buffer, and glucose in water, flows on the other. Despite all possible
modifications, cardiovascular death rate among HD patients is extraordinarily
high 3.       Peritoneal dialysis
involves the instillation of dialysate into the peritoneal cavity to allow
effective dialysis to occur across the membrane lining the abdominal cavity.
Peritoneal dialysis is continuous and fluid exchanges can either be performed
manually, usually four times a day (continuous ambulatory peritoneal dialysis,
CAPD) or can be done automatically overnight (automated peritoneal dialysis,
APD) which allows a greater number of fluid exchanges to be performed whilst
the patient sleeps. The frequency of fatal and non-fatal cardiovascular events is
increased even in the earliest stages of chronic kidney disease. 4-7 The analysis of USRDS database 8 extending
from 1977-1997 shows statistically significant alteration in HD group of
patients as compared to CAPD patients; with maximum death rates on Monday in HD
group and steady death rates were noted in patients using PD.      Jeloka et al 9 have shown that the cost of the
commonly prescribed HD and PD prescriptions are comparable. Fears of infection
due to the hot, humid climate and poor hygienic conditions have also turned out
to be unfounded; infection rates in most Indian PD programs are acceptable by
International standards. Still, PD is seldom offered as a first-choice dialysis
therapy, and only patients with multiple comorbidities not suitable for HD, are
initiated on PD.     Go et al determined
age standardized relative risk of cardiovascular events against eGFR (figure 1).
Those patients with CKD stages 4 or 5 have the highest risk 10. figure-1     This risk is even
higher within dialysis patients who display hugely elevated rates of cardiac
mortality, at least thirty-times greater than age matched controls 3      In the HEMO study,
the most common cause of death in dialyzed patients was ischemic heart disease
(20.4 %) followed by cardiac rhythm disorder (10.4 %), cerebrovascular disease
(8.6 %), and infections (7.7 %) 13      According to the
Annual Report 2001 of the United States Renal Data System (USRDS), the incidence
of new myocardial infarction in the first year of renal replacement therapy was
7.0 %, of cerebrovascular accidents 7.1 %, and of surgery for peripheral
vascular disease 8.4 % (Berger et al 1992).      As a crude
comparison, the Framingham Heart Study reported that the annual risk of
re-infarction in subjects surviving a first recognized myocardial infarction was
4 per cent per year, almost half the risk of new myocardial infarction in HD
patients 14 for whom the additional risk factor of uremia is compounded by
also being on dialysis.     Such high levels of
cardiovascular disease in dialysis patients are in part due to the high
prevalence of ‘traditional’ risk factors (diabetes, hypertension,
hyperlipidaemia, smoking, and physical inactivity). Unfortunately, aggressive
management of these traditional risk factors in dialysis patients has failed to
adequately control the progression of cardiovascular disease. For example, in uremic
patients, LV hypertrophy progresses with time on dialysis even when patients
are kept normotensive 15,16      In the study of
Parfrey, 71 % of non-diabetic dialysis patients without dilated cardiomyopathy
had LV hypertrophy and in the majority of patients this progressed over a
period of 3–4 years. Progression was not predicted by blood pressure, hyperparathyroidism
or anemia (although these factors are definitely involved), suggesting that
additional factors may play a part. Therefore this excess of cardiovascular
disease in dialysis patients must be explained by the presence of other unique
metabolic and hemodynamic derangements specific to the uremic patient – the so called
‘uremic’ specific risk factors. These uremic risk factors are less well defined
but are multiple, including some of those mentioned above already.      It is not only
systolic and diastolic blood pressures, which determine LV work and LV mass.
Vascular calcification with increased vessel stiffness is common, associated
with the development of left ventricular hypertrophy (LVH) 17 and also independently
predicts mortality 18     Stiffening of the
aorta and the associated increased impedance plays a major role in the
development of LVH 19. High pulse pressure is a surrogate marker for aortic
stiffness and pulse wave velocity is a potent predictor of LV mass and
cardiovascular events 20,21.     Diabetic patients
with ESRD have particularly high cardiovascular morbidity and mortality 22.
Even diabetic patients without nephropathy have major abnormalities of cardiac
structure. In observational studies, when compared to non-diabetic individuals,
diabetic patients have more severe LV hypertrophy and also develop ischemic
heart disease more frequently. 23     Anemia is a common
consequence of ESRD. Sustained anemia leads to vasodilatation, increased venous
return, cardiac enlargement and increased cardiac output. 24,25 Numerous
observational studies documented that anemia is associated with increased LV
mass in patients with CKD 26 and this is true even for apparently trivial
degrees of anemia. 27 Observational studies have also suggested that anemia is
an independent predictor of mortality. 28,30 Partial correction of anemia
partly corrects LV hypertrophy 30-32 but there is little or no controlled
evidence that reversal of anemia reduces cardiovascular mortality.      In the Canadian Normalization
of Hemoglobin Trial, hemodialysis patients with asymptomatic echocardiographic enlargement
were randomly treated to hemoglobin of 10 or 13.5 g/dl. The higher hemoglobin
failed to show a regression in left ventricular dilatation but subsequent
studies have shown that normal hemoglobin will prevent the development of new
LV dilatation. 33     Recurrent volume
overload with rapid fluid shifts (that occur on dialysis) increases cardiac
filling pressures and venous return, imposing an increased workload on the left
ventricle. Eventually this increase in preload results in LV dilatation and
LVH. A correlation is found between LV volume and blood volume. In patients
with fluid overload the heart diameters usually returns to normal a few hours
after ultrafiltration, but chronic overload may lead to eccentric LV hypertrophy
and irreversible dilatation.      The importance of hypervolemia
is illustrated by the observation of Ozkahaya et al. that volume control by low
salt diet and aggressive ultrafiltration reversed LV dilatation and hypertrophy
despite no administration of antihypertensive agents. 34,35     Cardiac mortality in
dialysis patients has also been linked to chronic inflammation, often manifested
as hypo albuminemia and elevated C-reactive protein (CRP) levels. Inflammation
leads to accelerated atherosclerosis, vascular calcification and increased
muscle catabolism. 36 Certainly, the elevation of several cytokines (in
particular CRP and interleukin-6) have also been shown to be associated with an
increase in mortality 37,38     The presence of LVH
itself predicts a worse long term outcome, and is associated with an increased
propensity to cardiac arrhythmias 39,40 LVH is more frequent in early stages
of chronic kidney disease and increases progressively, so that it is found in
approximately 70 per cent of patients starting renal replacement therapy. 41
LVH is not an innocent academic finding: Silberberg et al. had clearly
documented that it was an independent predictor of death on dialysis 42(figure-2)         (figure-2) shows that
LVH reduces survival in dialysis patients. Cumulative survival according to
echocardiographic LV hypertrophy defined as LV mitral inflow greater than 125
g/m2. This cut-off point corresponds to the 95th centile of the normal
population. 42     It has long been
suspected that myocardial ischemia may be precipitated by hemodialysis, with
the first evidence of silent ST segment depression during dialysis reporting
back to 1989. 43 However, this concept of dialysis induced subclinical ischemia
(occurring without acute atherosclerotic plaque rupture) has received
remarkably little attention, despite its theoretical plausibility. Short
intermittent hemodialysis treatments exert significant hemodynamic effects, and
20-30 per cent of treatments are complicated by intra-dialytic hypotension
(IDH). 44-46 In conjunction with this, hemodialysis patients are particularly
susceptible to myocardial ischemia. In addition to the high prevalence of
coronary artery atheroma 47,48 diabetic dialysis patients have been shown to
have a reduced coronary flow reserve (CFR) – the ability of coronary arteries
to dilate when myocardial demand is increased – even in the absence of coronary
vessel stenosis. 49     There is preliminary
evidence that the same phenomenon is also seen in non-diabetic dialysis
patients 50 which may be due to LVH, that leads to both structural and
functional reductions in the myocardial microcirculation.        The presence of LVH
on its own reduces coronary flow reserve (CFR) even in the absence of large
vessel coronary disease. 51 Because of the increased extravasal resistance,
coronary reserve is reduced as illustrated in the patient with aortic valve
stenosis. They suffer from angina pectoris resulting from ischemia despite
patent arteries. This is particularly important in HD patients in whom ischemia
is often asymptomatic. 52     Similarly the uremic
patient with LV hypertrophy may have ischemia intolerance when oxygen demand is
increased. In addition, the presence of concentric LVH renders the ventricle
more sensitive to acute changes in filling pressure, exactly as occurs during hemodialysis.
53 Increased peripheral artery stiffness is also recognized to have an
adverse effect on myocardial perfusion and reduces the ischemic threshold; 54
therefore, LVH in conjunction with increased vascular stiffness leads to a
propensity to reduced subendocardial blood flow. 55 Since the initial report
by Zuber et al 43, there have been further studies that have demonstrated
silent ST segment depression occurring during dialysis. 56-64 These studies
report the occurrence of dialysis induced ST depression at rates that vary
between 15 and 40 per cent. However, there has been ongoing debate as to
whether these electrocardiographic abnormalities reflect silent ischaemia or
changes in electrolyte concentrations. Other than this, there has been only one
subsequent study that has demonstrated ischaemia using an alternate technique.
Singh et al assessed dialysis induced ischaemia using sestamibi single photon emission
computed tomography (SPECT). 65 In an unselected group of ten dialysis
patients who were not known to have coronary artery disease, seven developed
perfusion defects during dialysis. Importantly, concurrent ST depression
occurred with the perfusion defects in only 3 patients, suggesting that electrocardiographic
assessment alone may underestimate the incidence of dialysis-induced ischemia.      It has previously
been demonstrated that myocardial stunning occurs as a direct consequence of hemodialysis
(and can be ameliorated by improving systemic hemodynamics whilst on treatment).
66,67 Therefore, if myocardial ischemia and stunning are induced by hemodialysis
then the process of hemodialysis itself, repeated three times a week, may
potentially contribute to chronic cardiac damage in this patient group.
Myocardial stunning is therefore increasingly thought to be an underappreciated
causative mechanism for heart failure in the hemodialysis population.      Altered renal
function in CKD itself deregulates cardio-renal axis causing cardiac remodeling
which results in a wide
spectrum of arrhythmias like supraventricular tachycardia, atrial fibrillation,
ventricular ectopic beats, ventricular tachyarrhythmia, and sudden cardiac
death (SCD). 6* Sudden
cardiac death appears to correlate with the peri-dialytic period. 68
Consequently, there have been a number of studies looking at the potential
pro-arrhythmogenic effects of HD, which reported the frequency of dialysis
induced arrhythmias as anywhere in the range of 5 – 75% of treatments. 69,70
The presence of both potentially life threatening complex ventricular
arrhythmias (CVAs) and premature ventricular complexes (PVCs) has been
associated with increased morbidity and mortality. CVAs (defined as Lown score
100* of 3 and above) are reported in up to 35% of HD patients during
treatment 56 and associated with negative prognostic factors including new
coronary events 104 and silent myocardial ischemia. 60 In the general
population, people with PVCs are more than twice as likely to die from coronary
artery disease (CAD)71 and their presence may play a vital role in assessment
of cardiovascular risk. The frequency and prevalence of intra- and
post-dialytic ventricular arrhythmias are higher in patients with significant
coronary stenosis; 72 suggesting that myocardial ischemia plays a direct role
in the induction and persistence of ventricular arrhythmias during and after
HD.      Myocardial ischemia is also known to lead
to the development of regional wall motion abnormalities (RWMAs) and myocardial
stunning as a direct consequence of HD. 73,74 This arises secondary to
ischemia caused by a reversible reduction in myocardial blood flow during the
HD treatment session even in the absence of significant CAD. 75 In addition
HD patients have been shown to reduce coronary flow reserves, 49 increased
prevalence of left ventricular hypertrophy and impaired microcirculation 76,
all of which predispose to demand myocardial ischemia. A number of therapeutic
strategies directly targeting the increased incidence of CVAs and PVCs in HD
patients have been utilized. Unfortunately the majority of these involve
additional pharmacological agents that may be poorly tolerated and have
undesirable side-effects. 77      No evidence currently
exists that myocardial ischemia may contribute to the development of both RWMAs
and ventricular arrhythmias. As ischemia induced RWMAs are potentially
preventable, 66,67 the identification of a common pathophysiological process
connecting both sudden and ischemic cardiac death in dialysis patients may
offer single therapeutic targets to reduce both causes of mortality. Intra-dialytic
hypotension (IDH) is a very serious clinical problem and remains a significant
cause of morbidity in the HD population, occurring in 20-30 % of treatments
78. In addition, a fall in blood pressure during dialysis predicts mortality.
77 Furthermore, IDH could potentially contribute to myocardial hypoperfusion
during dialysis. The main mechanism of IDH is rapid reduction of blood volume owing
to ultrafiltration and decrease in extracellular osmolarity during the dialysis
session. If the ultrafiltration rate exceeds the plasma refill rate, this will
lead to a reduction in circulating volume. Hypotension occurs when this
reduction in blood volume surpasses the compensatory mechanisms of the
cardiovascular system.     Various studies have shown that patients
on HD are more prone to life threatening arrhythmias as compared to patients on
PD. Hence ECG changes in these patients undergoing dialysis do have a
prognostic significance. Though several studies have investigated ECG changes
and arrhythmias in dialysis patients, very few compare the
differences in these changes in patients on HD vis-a-vis peritoneal dialysis.