Anda di halaman 1dari 11

5/8/13 1:09 AM

Starting Dialysis Is Dangerous

How Do We Balance the Risk?
Christopher W McIntyre, Steven J Rosansky Kidney Int. 2012;82(4):382-387.

Abstract and Introduction


Recent studies of timing of dialysis initiation have challenged the recent trend to earlier initiation of therapy. The observed outcomes though are a consequence of the balance between the risks of advanced uremia versus the inherent dangers relating to dialysis therapy itself. Many of these risks are inherent in how dialysis treatment is currently carried out, and may indeed be amenable to mitigation, through refinement of clinical practice (and potentially modality choice). This article aims to lay out a discussion relating to patient outcomes being the composite result of this balance, pivoting on the vulnerability of a particular patient to these attendant risks.

Dialysis treatment has long been accepted as a life-saving treatment in the setting of terminal uremia, and now offers lifesustaining treatment to approximately two million people around the world, in a wide variety of health-care systems. Despite the widespread adoption of this life-saving therapy, dialysis treatment (and in particular the initiation phase) is redolent with dangers, many of which are only recently starting to be fully appreciated. This inherent tension between benefit and harm has, up to this point, been explored mainly through a focus on timing of dialysis initiation. This area of interest has been twinned with further consideration of conservative nondialytic care in chronic kidney disease (CKD) stage 5 patients; detailed discussion of this literature, however, is beyond the scope of this particular article. Conventional wisdoms have evolved concerning the expected benefits of early dialysis initiation. These have largely led to very similar national and international guidelines regarding the best time to start dialytic treatment.[1] These guidelines have been associated with a marked increase over the past decade in patients to commencing dialysis at an ever earlier stage.[2] The core question as to when the best time to commence treatment, in both populations and individuals, has received remarkably little systematic study. There has been only a single randomized controlled trial examining the issue of early- versus late-start dialysis initiation.[3] The other investigations have almost entirely relied on retrospective interrogations of registry or other observational data sets. These studies have relied on statistical manipulations to adjust for case mix and comorbidity burden. More recent iterations of this approach have consistently identified earlier initiation as being associated with poorer outcomes in both general populations and subsets of younger patients without significant comorbidity. Since 2001, 11 observational studies have examined the issue of comorbidity-adjusted survival versus the estimated glomerular filtration rate (eGFR) level (as assessed by the serum creatinine-based Modification of Diet in Renal Disease equation) at dialysis initiation. All but two of these studies found a comorbidity-adjusted survival disadvantage of early dialysis initiation, possibly underestimated by 'lead time bias', and recently a variety of observational studies have reported a graded survival benefit associated with lower eGFR.[46] Comorbidity has been considered a crucial factor in many of the previous studies. LaSalle and co-workers[7] found earlystart disadvantage only in patients with the highest level of pre-dialysis comorbidity. Adjustment of survival for measured comorbidity, however, still seems to suggest no advantage in early start. However, comorbidity data collected at the Page 1 of 11

5/8/13 1:09 AM

registry level has not been well validated, with only one published study that found high specificity but low sensitivity of comorbidity data from the US Renal Data System. Although comorbidity is strongly associated with dialysis mortality, one study of patients with a low comorbid burden has also confirmed this trend to reduced survival in patients with higher starting eGFR.[4] The question of dialysis start has been the subject of a recently published randomized controlled trial.[3] Cooper and colleagues[3] randomly assigned 828 patients 18 years of age or older to early dialysis start estimated GFR by surface area corrected CockcroftGault equation of 1014 ml/min or late start estimated GFR of 57 ml/min. During a follow-up period of about 3.5 years, 152 of 404 patients in the early-start group (37.6%) and 155 of 424 in the late-start group (36.6%) died (hazard ratio with early initiation, 1.04; 95% confidence interval, 0.83 to 1.30; P=0.75). There was no significant difference between the groups in the frequency of adverse events (cardiovascular events, infections, or complications of dialysis) or in quality-of-life measures. There appeared to be no health economic benefit of an earlier start. However, it should be noted that a significant proportion of patients randomized to later start did in fact commence earlier, because of clinician assessment of need (and indeed the majority commenced peritoneal dialysis (PD), rather than hemodialysis (HD)). These data are consistent with the proposition that exposure to the harm associated with dialysis initiation overwhelms the survival advantage related to attempted correction of uremia in the setting of relatively less advanced CKD. This article introduces the concept that outcomes in dialysis are the composite result of negative effects of advanced CKD, balanced by the attendant risks of dialysis therapy, underpinned by the denominator of individual vulnerability of the patient to both of the above. This dynamic may be further influenced by attempts to mitigate many of those risks.

Potential Benefits of Dialysis Initiation

Although advanced CKD is associated with a plethora of metabolic, structural, and functional abnormalities, there is still very little certainty in mapping the individual abnormalities to specifics of patient symptoms or subsequent adverse events. This has strengthened (by default) the focus on nutrition, owing to the inability to robustly identify other surrogates for outcome that might be manipulated by the dialysis process. Malnutrition has been cited as the most common reason for clinicians choosing an earlier start (other than absolutes such as life-threatening hyperkalemia or oliguria-associated fluid overload),[8] and is contained in many guideline recommendations. This is largely driven by abundant observational data identifying low serum albumin as being the strongest determinant of mortality in dialysis patients. The NECOSAD study group suggested that increasing small-solute clearance (based on urea) might affect the observed pattern of developing malnutrition after dialysis start.[9] However, there are currently no studies that directly support the ability of conventional three-times-weekly dialysis to improve nutritional parameters, especially without enhanced nutritional supplements. No improvements were seen over 3 years of follow-up in the HEMO study of prevalent HD patients in either serum albumin or anthropomorphic measures of nutrition.[10] This study highlighted the importance of inflammation and its associated effects on serum albumin (rather than lower serum albumin being an exclusive or even primary marker of malnutrition). In a recent study by Rosansky and co-workers,[4] the increase in mortality risk propensity with earlier dialysis initiation was preserved throughout the range of serum albumin. Advanced CKD leads to a wide variety of subjective patient effects including pain, dyspnea, increasing fatigue, dependency, and depression. There are no prospective studies suggesting that initiation of current intermittent dialytic therapies consistently address this symptom burden. The application of more intensive dialysis regimes has been associated with an increase in many measures of nutrition and well-being.[11] However, at present only a small fraction of patients are undergoing this form of therapy, which is still at the early stages of rigorous scientific assessment.

Page 2 of 11

5/8/13 1:09 AM

Risks of Dialysis
Conventional dialysis treatment has many inherent risks for patients. This appears true in older frailer patients, and in those with minimal comorbidity. Some of these risks are fundamental to the dialytic therapy and some relate to management of the patient during the period of transition from nondialytic to dialytic management of end-stage renal disease. These risks may potentially be mitigated. The period of dialysis initiation is associated with a particular increase in patient dependency (with similar increment in mortality).[12] In studies looking at longer-term dialysis outcomes, the effects of these risks are not evenly distributed over the entire dialysis vintage, with an excess of mortality condensed into the first 612 months of therapy.[4] Survival in the first year of dialysis (in the United States) has decreased despite evidence of overall improvement in dialysis patient survival overall.[13] The window of opportunity for a number of these risks is often in the dialysis preparation period, with a reduced period for specialist care/preparation being associated with increased mortality in the dialytic phase of patient care.[14] Specific areas of risks that must be considered when initiating dialysis include the following:
Infection and Dialysis Accessrelated Issues

Risks of all forms of infection are many times higher once the patient is started on dialysis. This relates partially to uremicinduced immuno-incompetence, but also significantly to vascular access in HD patients. Catheter-related infections result either from migration of skin organisms along the catheter into the bloodstream or contamination and colonization of catheter lumens. Biofilm formation (formed from a combination of host and bacterial molecules) on the surface of catheters also has an important role in facilitating colonization and resistance to antibiotic therapy. Patients on dialysis are 100-fold more likely to develop methicillin-resistant Staphylococcus aureus septicemia than the general population and 800-fold more likely than if not dialyzed with a native arteriovenous fistula.[15] Exposure to hospital-acquired infection is often just a component of the myriad of dangers associated with the increased amount of in-patient care, which is often characteristic. Others might include exacerbation of malnutrition, hypostatic pneumonia, falls, and venous thromboembolism risk. Upper extremity native arteriovenous fistula is the vascular access of choice, supported by the Dialysis Outcome Quality Initiative (K/DOQI) of the National Kidney Foundation. The use of definitive vascular access in HD patients, rather than tunneled central venous catheters, is associated with sustained reduction in mortality.[16] This difference in survival has previously been entirely attributed to differences in access-related sepsis; however, it is being appreciated that arteriovenous fistula formation is associated with a series of structural and functional adaptations that reduce the propensity to demand ischemia and result in a lower-risk cardiovascular milieu (lower central blood pressure, reduced arterial stiffness, and improved ventricular contractile function),[17] and higher vascular access flows are associated with reduced acute recurrent HD-related cardiac injury (myocardial stunning).[18]
Decompensation of Nutrition and Psychosocial Problems

The 6-month interval around the first dialytic treatment may be one of mandated dietary restriction, depression, anxiety, and reduced appetite. These changes result in patients being exquisitely sensitive to worsening malnutrition. Patients newly started on dialysis face the physical and time requirements of their dialysis treatments (e.g., 4 h plus travelling time, 3 days a week for HD, or fluid exchanges 4 times every day for PD). They are also required to adhere to strict dietary restrictions and fluid restrictions despite having to take a median of 19 tablets a day.[19] Dialysis therapy and underlying advanced CKD are associated with lethargy, itching, pain dizziness on HD days, and bloating or gastrointestinal disturbance while on PD.[20] These factors combine to sap patients' confidence and interfere with activity, exercise, and social engagement. Depression and other psychosocial problems are often initiated or aggravated by dialysis start, and are often associated with fundamental reductions in a patient's ability to socialize, or to remain in chosen employment, Page 3 of 11

5/8/13 1:09 AM

amplifying these psychological stresses. Clinical depression is highly prevalent, affecting around 25% of patients,[21] and correlates to mortality risk.
Specific Dialysis-related Risks

Cardiovascular Effects Current conventional HD is less capable of exerting significant recurrent systemic circulatory stress that may be important in the development of cardiac disease, as well as perfusion-dependent injury of a wide range of vulnerable vascular beds. These include gut, brain, and potentially the kidney. This predominantly hemodynamic injury can therefore result in a mixed picture of direct perfusion-related injury, local/systemic inflammation, and potentiation of further cycles of injury.[22] Cardiac protection (and potentially other organ systems as well) may be possible by reducing this circulatory stress with modifications such as daily dialysis therapies to limit ultrafiltration requirements.[23] Cardiac arrhythmias and risk of sudden death appear to be the most important final fatal consequence in HD patients. A lengthening of the QT interval corrected for heart rate (QTc) predisposes to torsade de pointes ventricular tachycardia, and although many episodes may be self-terminating torsade can degenerate into ventricular fibrillation. In non-uremic patients, episodes of torsade de pointes often occur because of a combination of factors, including a long QTc in combination with bradycardia or electrolyte abnormalities such as hypokalemia, hypomagnesemia, or hypocalcemia. Several authors have reported an acute increase in both QTc and QTd following dialysis. Although this would support the increase in sudden death temporally related to the dialysis procedure,[24] there are relatively few data showing a direct link between increased QTc or QTd induced by dialysis and either arrhythmia or sudden death. Nakamura et al. [25] compared two groups of 24 dialysis patients separated into those who displayed an increase in QTc post dialysis and those who did not. They reported an increase in cardiac events and cardiac mortality in the group with an increase in QTc post dialysis, but many of the outcomes were not arrhythmias or sudden death.[25] Conversely, two studies have documented the increase in QTc with dialysis but found that there was no resultant increase in arrhythmias.[26,27] There is also a lack of consensus regarding the mechanisms underlying the acute increase in QTc and QTd with dialysis. It would seem logical that shifts in plasma electrolytes would be the main cause. In essence, the degree of reduction and the absolute end-dialysis plasma concentrations of potassium, calcium, magnesium, and pH have all been linked to lengthening QTc and an increase in arrhythmias, whereas other authors have not found such associations. The longer interdialytic interval in three-times-weekly dialysis schedules and the use of lower dialysate potassium have been linked to increased risk of cardiac arrest in HD patients. Residual Renal Function The maintenance of residual renal function (RRF) is one of the key factors that has been identified to associate with better survival in HD patients.[28] An accelerated decline in RRF is, however, a characteristic of initiation of conventional three-times-weekly dialysis. This results in both a reduction in urine volume (with exacerbation of long term fluid control and increased ultrafiltration requirements during each HD session) and a reduction in the overall level of clearance of uremic toxins (in particular middle molecule solutes). Previously we have demonstrated that conventional HD has significant intradialytic hemodynamic effects. We hypothesize that the renal circulation is also vulnerable and that the same processes may contribute to new recurrent renal ischemic injury, driving the loss of RRF. This creates a vicious cycle with falling RRF, resulting in increasing interdialytic fluid gains and yet higher ultrafiltration requirements, driving additional injury to vulnerable vascular beds. Dialysis-associated loss of RRF is also associated with an increase in sudden cardiac death.[29] A detailed discussion of potential therapeutic approaches to help preserve RRF is beyond the scope of this article but might include the following: careful avoidance of dehydration through diuretics or excessive ultrafiltration, consideration of PD in preference to HD, and ensuring the use of only biocompatible dialysis membranes and possibly PD fluids. One potential approach to mitigate some of the risks associated with dialysis initiation might be to adopt the incremental approach, adding in additional HD time or sessions, or even a planned early use of PD with transition to HD, with Page 4 of 11

5/8/13 1:09 AM

subsequent waning of RRF. These approaches, although apparently reasonable, have not been subjected to rigorous study. Early use of even more intermittent short HD runs risks of exacerbated circulatory stress, and potential for further reduction in RRF. Tracking the appropriate stepping up of delivered dialysis dose is also challenging, especially at dialysis initiation with such profound changes in body composition and cardiovascular status. Additional use of PD also has inherent risks of the complications of both therapies, both in terms of access and cumulative metabolic stress. However, PD and HD have been used in a combined approach[30] in patients failing on either initial solo modality or even de novo at the start of renal replacement therapy.[31] Peritoneal Dialysis versus Hemodialysis Compared with nondialysis patients, both HD and PD patients have an 8-fold increase in cardiovascular risk and non-cardiovascular risks of death. The pathophysiological mechanisms that result in this high death rate are not well understood and may be different in HD versus PD patients. In HD patients, cardiovascular insults appear to be predominately hemodynamic and may involve biocompatibility issues of dialyzers used. PD results in hemodynamic effects, but the response associated with a low ultrafiltration rate is not characteristically hypotensive and does not result in acute cardiac injury.[32] In both dialytic modalities, the changes may be related to fluid type and glucose exposure. Metabolic insults may drive cardiovascular structural and functional pathophysiological processes. Deposition of advanced glycation end products in vascular tissues may result in reduced compliance of arteries and ventricular muscle.[33] Both short-term and longer-term metabolic factors may alter autonomic reactivity. All of these changes may prime patients to experience aberrant regional blood flow to vital organ systems. This principally hemodynamic versus metabolic comparison is superimposed on differing patterns of infection between the modalities and specific risks associated with peritoneal malfunction and failure. One potential approach to mitigate some of the risks associated with dialysis initiation might be to adopt the incremental approach, adding in additional HD time or sessions, or even a planned early use of PD with transition to HD, with subsequent waning of RRF. These approaches, although apparently reasonable, have not been subjected to rigorous study. Early use of even more intermittent short HD runs risks of exacerbated circulatory stress, and potential for further reduction in RRF. Tracking the appropriate stepping up of delivered dialysis dose is also challenging, especially at dialysis initiation with such profound changes in body composition and cardiovascular status. Additional use of PD also has inherent risks of the complications of both therapies, both in terms of access and cumulative metabolic stress. However, PD and HD have been used in a combined approach in patients failing on either initial solo modality or even de novo at the start of renal replacement therapy.

Conflation of Risk and Susceptibility: Future Directions of Study

Patients requiring dialysis are subject to significant functional and structural abnormalities, which may sensitize them to conventional mortality risks, as well as create other issues that are unique to dialytic therapy. These drivers of mortality may relate to end-stage CKD, dialytic therapy, and an individual's susceptibility to both (Figure 1 and ). To date, no attempt has been made to assess the relative effects of all of these risks in a particular patient situation. Research up to this point has focused on thresholds, failing to embrace the reality of biology being essentially a continuum (and therefore lacking biological plausibility). It also has been predicated on there being more risk inherent in having advanced uramia than that being faced associated with renal replacement therapy.
Table 1. Summary of the attendant recognized risks of dialysis

HD Risks at initiation Increasing dietary restriction Escalation of medication burden


Page 5 of 11

5/8/13 1:09 AM

Specific acute hospitalization risks

Venous thromboembolism Hospital-acquired infection

Dialysis access

CRI AVF/arteriovenous graft as portal of sepsis

Peritonitis Exit site infection

Overall infection risk

Pneumonia Generalized increased bacteremia risk


Increasing dependency Anxiety and depression

Dialysis specific

Circulatory stress Cardiac injury Brain ischemic injury Increased bleeding risk

Acute and chronic metabolic stress


Subclinical access infection Biocompatibility issues Translocation of endotoxin from gut due to congestion and enteral ischemia

Sudden cardiac death

Hemodynamic/ischemic stress of HD Electrolyte disturbance

Loss of residual renal function


Suppressed appetite Disturbed catabolic/anabolic balance Accelerated muscle wasting

Peritoneal protein losses Appetite suppression by intra-abdominal volume and hypertonic glucose fluids

Loss of residual renal function

Biocompatibility HD-related recurrent acute kidney injury

Inappropriate low target weight and inappropriate ultrafiltration

Abbreviations: AVF, arteriovenous fistula; CRI, catheter-related infection; HD, hemodialysis; PD, peritoneal dialysis. HD and PD are considered separately to highlight that some of this risk profile is shared and some is far more dialysis modality specific.

Page 6 of 11

5/8/13 1:09 AM

Figure 1.

Chances of survival at any given starting glomerular filtration rate are a complex dynamic balance between the benefits of relieving advanced uremia and the risks of the attendant therapy. The pivot point for this balance is the individual patient tolerability of both the uremia and the proposed renal replacement therapy. This pivot point itself is somewhat different between patients and even within the same patient over a period of time. The relative weight of the risk of being markedly uremic may well be substantially less than those inherent in dialysis (as it is commonly initiated and performed). Although such an approach might seem difficult, it is not impossible. Risks from therapies can be quantified, such as with the use of advanced real-time functional assessments (tissue perfusion, systemic hemodynamic responses, cardiac responses to dialysis treatment, and others), which can define the response of a healthy human to a particular level of Page 7 of 11

5/8/13 1:09 AM

dialytic stress.[34] It may also be possible to focus on biological factors that are acting as integrators of these susceptibilities. Potential candidates might be humoral (markers of tissue injury, inflammation, malnutrition, and others), whole-body functioning (6-min walk test, sit to stand testing, and others), health-related quality of life (with specific testing of disease-related domains and so on), or body composition (measuring hydration or reduction in lean body mass and so on). As an example, muscle mass can be accurately assessed by measuring muscle cross-sectional area (MCSA) on thigh CT. There are many factors already reported as being associated on a cross-sectional study basis with muscle wasting. These include decreasing glomerular filtration rate, dialysis, age, and diabetes.[35,36] Muscle atrophy can be assessed by various methods, and previous work has shown good correlation between thigh MCSA and functional performance, serum albumin, age, and inflammatory status in CKD stages 45, HD, and PD patients. These changes have previously been attributed to inadequate dietary protein intake, but recent advances in our understanding of this area indicate that other processes are involved. There is a prospectively observed significant loss of muscle CSA in late-stage CKD 4, suggesting that these patients are subject to increased disturbance of the catabolic/anabolic balance, presumably as a result of advancing uremia. This occurs at eGFRs between 10 and 20 ml/min.[37] Dialysis initiation based purely on estimation of renal function from serum creatinine runs the danger of allowing patients to 'loose ground' in terms of overall nutritional status in these late stages, and failing to optimally time dialysis initiation. This raises the possibility that we might be able to utilize this acceleration of muscle wasting as an individualized biomarker of increasing metabolic decompensation, providing a rationale for when the appropriate time to start dialysis in an individual might be. An alternative approach might be to identify those patients at particular risk of the cardiovascular effects of dialysis. An approach based on the detection of large-vessel coronary artery disease is unlikely to be useful, given the relatively reduced significance of such findings in the dialysis population. Appropriate cardiovascular stress studies, more sensitive measures of reduced ventricular function (such as global longitudinal strain), microcirculatory studies, or more advanced analyses of heart rate variability patterns (providing information on autonomic function) all might hold promise. The consequence, however, would potentially be to defer dialysis initiation in those patients with documented patterns of cardiovascular vulnerability. Application of such a strategy might see our frailest and most vulnerable patients allowed to be exposed to worsening uremia, and worsened overall condition when they did commence treatment. The risk/benefit ratio for these patients (Figure 1) may be such that nondialytic conservative management may be the wisest choice.

There is still considerable doubt concerning the optimal timing of dialysis initiation. This is true for general populations, especially patient groups (elderly, diabetics) and individuals with unique susceptibilities to adverse outcomes with dialysis initiation. Preliminary randomized controlled trial and registry-based data have challenged conventional wisdom that an earlier start is beneficial. To date, none of the studies have considered the interaction between uremic need, dialysisrelated harm, and individual patient susceptibility to these factors. This balance is very important, as currently we have limited ability to establish the individual vulnerability to either advancing CKD or dialysis initiation. The degree to which the 'pivot' of this balance may be changed by addressing elements of the patients' particular vulnerability or the weight of dialysis-associated risk is not yet fully understood. Although direct testing of this central hypothesis (relating to the balance of risk and vulnerability) is challenging, consideration initially aimed at quantifying these individual risks and integrating them to provide coherent choices for initiation is a crucial priority to improve patient outcomes. Such an enhanced appreciation of the biological realities faced by our patients would allow the utilization of the expensive and potentially harmful renal replacement resource in the most effective manner.

1. Rosansky S, Glassock RJ, Clark WF. Early start of dialysis: a critical review. Clin J Am Soc Nephrol 2011; 6: 1222 Page 8 of 11

5/8/13 1:09 AM

1228. 2. US Renal Data System. USRDS 2009 Annual Data Report. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases: Bethesda, 2009. 3. Cooper BA, Branley P, Bulfone L, for the IDEAL Study et al. A randomized controlled trial of early versus late initiation of dialysis. N Engl J Med 2010; 363: 606619. 4. Rosansky SJ, Eggers P, Jackson K et al. Early start of hemodialysis may be harmful. Arch Intern Med 2011; 171: 396403. 5. Clark WF, Na Y, Rosansky SJ et al. Estimated glomerular filtration rate at dialysis initiation and mortality. CMAJ 2011; 183: 4753. 6. Hwang S, Yang W, Lin M, Taiwan Society of Nephrology et al. Impact of the clinical conditions at dialysis initiation on mortality in incident haemodialysis patients' national cohort study in Taiwan. Nephrol Dial Transplant 2010; 25: 26162624. 7. LaSalle M, Labeeuw M, Frimat L, on behalf of the REIN Registry et al. Age and comorbidity explain the paradoxical association of early dialysis starts with poor survival. Kidney Int 2010; 77: 700707. 8. Ledebo I, Kessler M, van Biesen W et al. Initiation of dialysisopinions from an international survey: report on the dialysis opinion symposium at the ERA-EDTA Congress, 18 September 2000, Nice. Nephrol Dial Transplant 2001; 16: 11321138. 9. Jansen MA, Korevaar JC, Dekker FW, NECOSAD Study Group et al. Renal function and nutritional status at the start of chronic dialysis treatment. J Am Soc Nephrol 2001; 12: 157163. 10. Friedman AN, Fadem SZ. Reassessment of albumin as a nutritional marker in kidney disease. J Am Soc Nephrol 2010; 21: 223230. 11. Chertow GM, Levin NW, Beck GJ et al. In-center hemodialysis six times per week versus three times per week. N Engl J Med 2010; 363: 22872300. 12. Kurella Tamura M, Covinsky KE, Chertow GM et al. Functional status of elderly adults before and after initiation of dialysis. N Engl J Med 2009; 361: 15391547. 13. Collins AJ, Foley RN, Gilbertson DT et al. The state of chronic kidney disease, ESRD, and morbidity and mortality in the first year of dialysis. Clin J Am Soc Nephrol 2009; 4(Suppl 1): S511. 14. Kessler M, Frimat L, Panescu V et al. Impact of nephrology referral on early and midterm outcomes in ESRD: EPidmiologie de l'Insuffisance REnale chronique terminale en Lorraine (EPIREL): results of a 2-year, prospective, community-based study. Am J Kidney Dis 2003; 42: 474485. 15. Fluck RJ. Summary of the 5th edition of the Renal Association Clinical Practice Guidelines (2009-2012). Nephron Clin Pract 2011; 118(Suppl 1): c27c70. 16. Polkinghorne KR, McDonald SP, Atkins RC et al. Vascular access and all-cause mortality: a propensity score analysis. J Am Soc Nephrol 2004; 15: 477486.

Page 9 of 11

5/8/13 1:09 AM

17. Korsheed S, Eldehni MT, John SG et al. Effects of arteriovenous fistula formation on arterial stiffness and cardiovascular performance and function. Nephrol Dial Transplant 2011; 26: 32963302. 18. Korsheed S, Burton JO, McIntyre CW. Higher arteriovenous fistulae blood flows are associated with a lower level of dialysis-induced cardiac injury. Hemodial Int 2009; 13: 505511. 19. Chiu YW. Pill burden, adherence, hyperphosphatemia, and quality of life in maintenance dialysis patients. Clin J Am Soc Nephrol 2009; 4: 10891096. 20. Weisbord SD. Prevalence, severity, and importance of physical and emotional symptoms in chronic hemodialysis patients. J Am Soc Nephrol 2005; 16: 24872494. 21. Cohen SD. Screening, diagnosis, and treatment of depression in patients with end-stage renal disease. Clin J Am Soc Nephrol 2007; 2: 13321342. 22. McIntyre CW. Recurrent circulatory stress: the dark side of dialysis. Semin Dial 2010; 23: 449451. 23. Jefferies HJ, Virk B, Schiller B et al. Frequent hemodialysis schedules are associated with reduced levels of dialysis-induced cardiac injury (myocardial stunning). Clin J Am Soc Nephrol 2011; 6: 13261332. 24. Bleyer AJ, Hartman J, Brannon PC et al. Characteristics of sudden death in hemodialysis patients. Kidney Int 2006; 69: 22682273. 25. Nakamura S, Ogata C, Aihara N et al. QTc dispersion in haemodialysis patients with cardiac complications. Nephrology 2005; 10: 113118. 26. Kayatas M, Erturk A, Muderrisoglu H et al. Acetate hemodialysis does not increase the frequency of arrhythmia in hemodialysis patients. Artifi Organs 1998; 22: 781784. 27. Korkmaz M, Muderrisoglu H, Ozdemir N et al. QT interval lengthening during haemodialysis: is there a potential risk for torsade de pointes? Nephrol Dial Transplant 1997; 12: 365. 28. Shemin D, Bostom AG, Laliberty P et al. Residual renal function and mortality risk in hemodialysis patients. Am J Kidney Dis 2001; 38: 8590. 29. Pun PH, Smarz TR, Honeycutt EF, for the AASK Collaborative Research Group et al. Chronic kidney disease associated with increased risk of sudden cardiac death among patients with coronary artery disease. Kidney Int 2009; 76: 652658. 30. Kawanishi H, McIntyre C. Complementary use of peritoneal and hemodialysis: therapeutic synergies in the treatment of end-stage renal failure patients. Kidney Int Suppl 2008; 108: S63S67. 31. McIntyre CW. Bimodal dialysis: an integrated approach to renal replacement therapy. Perit Dial Int 2004; 24: 547 553. 32. Selby NM, McIntyre CW. Peritoneal dialysis is not associated with myocardial stunning. Perit Dial Int 2011; 31: 27 33. 33. McIntyre NJ, Chesterton LJ, John SG et al. Tissue-advanced glycation end product concentration in dialysis patients. Clin J Am Soc Nephrol 2010; 5: 5155. Page 10 of 11

5/8/13 1:09 AM

34. McIntyre CW, Odudu A, Eldehni MT. Cardiac assessment in chronic kidney disease. Curr Opin Nephrol Hypertens 2009; 18: 501506. 35. Ohkawa S, Odamaki M, Ikegaya N et al. Association of age with muscle mass, fat mass and fat distribution in nondiabetic haemodialysis patients. Nephrol Dial Transplant 2005; 20: 945949. 36. McIntyre CW, Selby NM, Sigrist M et al. Patients receiving maintenance dialysis have more severe functionally significant skeletal muscle wasting than patients with dialysis-independent chronic kidney disease. Nephrol Dial Transplant 2006; 21: 22102216. 37. John SG, Sigrist MK, McIntyre CW. Natural history of skeletal muscle mass changes in CKD stage 4 and 5 patients. J Am Soc Nephrol 2007; 18: 768A (P082).

Kidney Int. 2012;82(4):382-387. 2012 Nature Publishing Group

Page 11 of 11