Dialysis-induced segmental wall motion abnormalities, post-dialysis fatigue and cardiovascular mortality: the new Bermuda triangle?

Dialysis patients exhibit higher rates of death compared with the general population [1], with an almost similar increase in the proportion of cardiovascular (CV) and non-CV deaths [1]; however, the CV mortality risk in patients starting dialysis is much higher compared with the general population (unstandardized CV risk—15-fold higher/age-standardized CV risk—8.8 higher than in matched individuals from the general population) [1] and is not determined by the same risk factors. Indeed, classical atherosclerotic disease is not the most important and common cause of death in a modern dialysis population. Sudden cardiac death (SCD) [2] and heart failure are the most frequent forms of death in dialysis patients; the overall incidence of these events is much greater than the incidence of coronary events [3]. Moreover, the risk persists after coronary vascularization [4]. Important factors for this significant CV morbidity include left ventricular (LV) hypertrophy, electrolyte shifts during haemodialysis (HD), poor volume control [5] and abnormalities in myocardial ultra-structure and function, including endothelial dysfunction, interstitial fibrosis, impaired coronary flow reserve and diminished ischaemia tolerance. One potential important player in the mechanisms behind this excess mortality is the HD procedure. It is obvious that the HD procedure is stressful for the CV system; almost 30% of HD sessions are associated with significant intra-dialytic hypotension [6]. Intermittent HD treatments may induce global and segmental myocardial ischaemia [7, 8]. This was suggested by Selby's studies using echocardiography [9, 10] and cardiac troponin levels [11] and by Singh et al. [12] using 99Tc-sestamibi single-photon emission computed tomography. Confirmation that an HD session is undeniably capable of inducing regional myocardial ischaemia was provided recently by McIntyre et al. [7]. Using intra-dialytic H215O positron emission tomography and echocardiography, they showed that HD treatment elicited reversible (LV) regional wall motion abnormalities (RWMAs) with matched reductions in myocardial blood flow (MBF). Dasselaar et al. [8] reconfirmed 1 year later these results in non-diabetic non-cardiac-compromised patients. They observed a significant reduction in MBF, early during dialyses that were not accompanied by significant fluid removal. In the general population, repeated episodes of transient myocardial ischaemia lead to a spectrum of disease encompassing myocardial stunning, myocardial hibernation and ending in myocardial remodelling and scarring, with irreversible loss of contractile function. Myocardial stunning is defined as a prolonged LV dysfunction without histological signs of irreversible injury to cardiomyocytes. It occurs each time when ischaemia and reperfusion are present: in patients with coronary artery disease, after percutaneous transluminal coronary angioplasty, after coronary bypass or cardiac transplantation or following resuscitation. Myocardial stunning has been well described in the general population as a causative mechanism for heart failure [9]. Myocardial hibernation is described also as depressed myocardial contractility, but the ischaemic myocardium remains viable and the LV dysfunction can be partially or completely restored to normality, by improving/restoring blood flow. In pathophysiological terms, a severe reduction in coronary flow reserve is common for both conditions, and recovery of function is paralleled by restoration of an adequate coronary flow reserve. The reduction in MBF during HD is associated with myocardial remodelling and eventually non-viable myocardium with irreversible reduction in LV function. This process becomes increasingly appreciated as a principal pathophysiologic contributor to the excess CV mortality in HD patients [13]. Repeated episodes of myocardial ischaemia determine a spectrum of disease, from segmental myocardial stunning to myocardial hibernation, resulting in myocardial remodelling and scarring and irreversible loss of contractile function [7]. This process becomes increasingly appreciated as a principal pathophysiologic contributor to the excess CV mortality in HD patients [11]. In a recent study in patients without coronary artery disease (excluded after coronary angiography), the HD session induced a reduction in MBF to levels consistent with the development of overt myocardial ischaemia [5]. Moreover, this observed reduction in MBF correlated directly with the occurrence of segmental LV dysfunction as measured by simultaneous echocardiography [5]. After a 12-month period of follow-up, this dialysis-induced RWMA progressed to LV failure with increased risk for ventricular arrhythmias [12]. As expected, the mortality rate was higher at 12 months in patients who had myocardial stunning [13]. Although the effects of myocardial regional ischaemia on cardiac function and general mortality in dialysis patients are studied, the associations of myocardial regional ischaemia with clinical symptoms have not been investigated yet. In the general population, the literature describes a continuum spectrum of clinical symptoms in patients with stress induced myocardial disease, ranging from mild subclinical variants (including cadiac fatigue) to the severely symptomatic patient presentation. Cardiac fatigue was described after prolonged exercise. Starnes and Bowles suggested that transient ischaemia was a likely mechanism for cardiac fatigue and could result from a process called myocardial stunning [13]. Cardiac fatigue as a result of transient ischaemia during exercise is associated with the accumulation of oxygen-free radicals and alterations in calcium metabolism [14]. ‘Dubin et al. in this issue of NDT suggests that severe post dialysis fatigue (PDF) could be a clinical manifestation of cardiac ischaemia occurring during dialysis’. Could we speculate that PDF is a form of cardiac fatigue secondary to repeated episodes of myocardial ischaemia during HD? Dubin et al. recruited and studied 40 patients on a stable chronic HD regimen; all of them had repeated echocardiographic examinations (before and during the last hour of the second or third dialysis session of the week) for at least two dialysis sessions and also completed questionnaires about their symptoms. WMA was measured on these echocardiograms and scored by a blinded reviewer (myocardial segments were graded 1–4, with 4 representing the worst WMA and the segmental scores were summed for each echocardiogram). Subsequently, the change in WMA score was calculated: a positive value indicated worsening of the LV wall motion, while negative values represented improvement in WMA. Severe PDF (defined as lasting more than 2 h after dialysis) was analysed using a generalized linear model with candidate predictors including anaemia, intra-dialytic haemodynamics and cardiac function. The authors found that 44% of patients with worsening WMA had severe PDF, compared with only 13% of patients with improved or unchanged WMA (P = 0.04). Furthermore, one-point increase in WMA score during dialysis was associated with a 10% higher RR of severe PDF. After multivariable adjustment, every point increase in WMA score was still associated with a nearly doubled relative risk of severe PDF, independent of demographics, co-morbidities and haemodynamics. Although the cross-sectional design did not allowed to establish a direct causal relationship, the data raise the provocative possibility that, in some cases, severe PDF may be a sequela of intra-dialytic ischaemia. Moreover, the authors speculate that slow vascular refilling causes reduced perfusion to both the brain and the heart for hours following dialysis, causing both centrally mediated fatigue and reduced myocardial perfusion. Fatigue is one of the most common symptoms experienced by dialysis patients, with its prevalence ranging from 60 to 97% [15]. PDF occurs often after dialysis sessions and is associated with limitation of the patient's functional independence and participation in social activities following dialysis [16]. Moreover, patients need almost 5 h of sleep to recover after their HD session and had more depression and insomnia. In addition to impaired physical and social functioning, fatigue has been associated with a lower quality of life and premature death in patients on HD. Fatigue is an important predictor for CV events, independent of traditional risk factors, including age, presence of diabetes and CVD history [17] or non-traditional risk factors, such as malnutrition or inflammation. More severe fatigue at baseline was independently associated with an increased risk of all-cause mortality, even after adjusting for traditional risk factors [18]. Moreover, the longitudinal changes in fatigue over time are also independently associated with mortality in HD patients; every 10-point increase in the vitality score over time increased the chances of survival by 10% [19]. The mechanism is not fully understood. Abnormal lipids metabolism, decreased fibrinolysis, low heart-rate variability and higher levels of pro-inflammatory cytokines associated with fatigue could determine such an increased CV and general risk in HD population [18]. A great variety of strategies are used in observational studies to decrease the risk of myocardial stunning, but randomized controlled trial are lacking. A modified dialysis regimen such as biofeedback dialysis, cooling of the dialysate and increased dialysis frequency has recently been shown to reduce HD-induced stunning over standard thrice-weekly dialysis. Recently, Jefferies et al. [20] described the benefit of more frequent dialysis (five to six times weekly, home dialysis and home nocturnal dialysis) on LV function and regional wall motion. The biofeedback technique was compared by McIntyre's group with standard HD: there was a 1.8-fold increased risk of myocardial stunning in the standard HD group [10]. By cooling dialysate from conventional 37°C to patient-individualized body temperature significantly reduces the number of RWMAs per patient [21]. Although promising, the results of Dubin's study must be taken with great prudence. PDF was assessed only by a questionnaire covering the last 30 days, and was defined solely on the basis of duration rather than intensity. Furthermore, the cohort is small (N = 40), and contains a high proportion of Afro-Americans (38%)—is this aspect the major determinant of the small proportion of WMA (only 26% compared with almost 60% in other studies) [7]? Is myocardial stunning during dialysis really less common in non-whites and if is so, could it contribute to the survival advantage of African-American dialysis patients? More importantly, could these results be generalized to other end-stage renal disease cohorts? Further longitudinal studies are needed to clarify the relationship between WMA and PDF and whether therapies aimed at WMA could reduce PDF and improve quality of life and survival. Clinical symptoms would be extremely important in identifying high-risk patients and to optimize their management in order to reduce myocardial stunning and CV morbidity and mortality. None declared. (See related article by Dubin et al. Association of segmental wall motion abnormalities occurring during hemodialysis with post-dialysis fatigue. Nephrol Dial Transplant 2013; 28: 2580–2585.)