"Timing of Dialysis Inititiation and Choice of Dialysis Modality":
http://www.medscape.com/viewarticle/740184
That link didn't open for me but I found this:
Nature Reviews Nephrology 7, 66-68 (February 2011) | doi:10.1038/nrneph.2010.178
Subject Category: Dialysis (hemodialysis, peritoneal dialysis, continuous renal replacement)
End-stage renal disease in 2010: Timing of dialysis initiation and choice of dialysis modalityBengt Lindholm & Simon Davies About the authors
Abstract
Few, if any, doubt that renal transplantation should be the first choice of renal replacement therapy whenever this option is possible. But, at which level of remaining kidney function should patients start on dialysis and which dialysis modality should be used? Several studies published in 2010 addressed these questions and will influence dialysis initiation strategies worldwide for years to come.
Data from the US Renal Data System (USRDS) between 1996 and 2008 showed that the proportion of patients initiating hemodialysis with an estimated glomerular filtration rate (eGFR) >10 ml/min/1.73 m2 increased from 20% to 52% and those with a starting eGFR of ≥15 ml/min/1.73 m2 increased from 4% to 17%.1, 2 Without evidence of beneficial effects, this early initiation of dialysis incurs treatment-associated risks, patient lifestyle restrictions and substantial medical costs.
In this light, the Initiating Dialysis Early and Late (IDEAL) study, which was conducted in Australia and New Zealand and showed that early initiation of dialysis had no significant effect on the rate of death from any cause or on cardiovascular, infectious or other complications of dialysis itself, is timely.3 Designed to determine whether early versus late start reduces the rate of death, patients were randomly assigned to planned initiation of dialysis when their eGFR was 10–14 ml/min/1.73 m2 (early start) or when their eGFR was 5–7 ml/min/1.73 m2 (late start). Commencing in 2000, 828 patients were enrolled over 8 years; mean difference between the groups was just under 6 months between recruitment and initiation of dialysis. Of note, 75.9% of the patients in the late-start group initiated dialysis above the target eGFR owing to symptom development. During a median follow-up of 3.59 years, 37.6% of early starters and 36.6% of late starters died (hazard ratio
for early initiation 1.04, 95% CI 0.83–1.30; P = 0.75). The authors concluded that with careful clinical management, dialysis can be delayed for some patients until eGFR drops below 7 ml/min/1.73 m2 or until more traditional clinical indicators are present.
Although randomized trials provide the best type of evidence, a question always remains over the generalizability of their results. Several recent observational analyses have suggested that initiating dialysis at higher eGFRs is associated with increased mortality risk;1, 2 results from the Taiwanese national cohort study are an important addition to these data as the study explored an even lower range of initiating kidney function.4 This study demonstrated an inverse association between starting eGFR and survival rate among 23,551 incident hemodialysis patients with a median eGFR at dialysis initiation of 4.7 ml/min/1.73 m2, average age 61.5 years, and a remarkably low rate of mortality during the first year of dialysis (13.2/100 patient-years). In an attempt to adjust for case mix, a propensity score analysis was performed and also showed higher eGFR at dialysis initiation to be associated with increased mortality risk.
© iStockphoto
Care should be taken in interpreting observational studies of this kind, partly because of the possibility of unmeasured—and thus unadjusted—risks in patients with higher eGFR (for example, disproportional loss in muscle mass), but also because of the survivor effect. It could be that patients with lower eGFRs are 'selected' as survivors before dialysis initiation is even considered, leading to an informative censoring bias. Two other studies designed to address these concerns were published in 2010.2, 5
In a population-based, prospective, observational cohort study of all Swedish citizens with a certain degree (mean eGFR 16.1 ml/min/1.73 m2) of chronic kidney disease (CKD) identified between 1996 and 1998, 901 patients were followed for 5–7 years.5 By June 2003, 736 patients had started renal replacement therapy (RRT; mean eGFR 7.6 ml/min/1.73 m2; 309 of whom died), 90 had died before any RRT was initiated and 56 were still alive without RRT. Again, initiating dialysis at a below-median eGFR was associated with a reduced risk of death (HR 0.84, 95% CI 0.64–1.10), whereas mortality rate in never-dialyzed patients with an eGFR <7.5 ml/min/1.73 m2 was greatly increased (compared with never-dialyzed patients with eGFR 7.5–10 ml/min/1.73 m2; HR 4.65, 95% CI 2.28–9.49), suggesting that some survivor bias was indeed present in this patient group.
Conversely, if the observed increased mortality associated with a higher eGFR at initiation is caused by worse clinical condition of these patients it might be expected that the adverse effect of starting earlier would be attenuated in patients without evidence of major comorbidity. To test this hypothesis, outcomes of 81,176 patients commencing in-center hemodialysis in the US, selected for lack of comorbidity by excluding patients with diabetes and those aged ≥65 years, were analyzed by starting eGFR and category of plasma albumin—a surrogate for unmeasured morbidity.2 In contrast to the prediction, starting eGFR had little effect on survival of the sicker patients whereas among fitter patients the higher the starting eGFR, the greater the relative increase in mortality (HRs 1.53 and 2.18 for eGFRs of 10–15 ml/min/1.73 m2 and >15 ml/min/1.73 m2, respectively). The authors suggest that early start of hemodialysis may in fact be harmful, questioning the trend to early hemodialysis initiation and proposing that initiation of hemodialysis should not be based on an arbitrary level of eGFR or on serum creatinine level unless accompanied by other definitive indications.
The whole idea of dependency on this single metric [eGFR] may in fact be flawed...
2010 saw publication of two further studies comparing survival of patients in the US commencing either in-center hemodialysis or peritoneal dialysis (PD). Both used the USRDS and found that overall, survival did not differ by modality. One study, which included 620,020 hemodialysis patients and 64,406 PD patients from all US centers, demonstrated that PD outcomes steadily improved between 1996 and 2004, eliminating the previously observed increased risk in starting with PD.6 This finding was seen in subgroups defined by age, diabetes and additional comorbidities, and survival was equivalent over 5 years of follow-up.
The second study undertook a detailed propensity-matched analysis approach to account as much as possible for unmeasured modality selection effects.7 Based on 6,337 hemodialysis–PD patient pairs from a retrospective cohort of 98,875 adults initiating dialysis in the US in 2003, an intention-to-treat analysis of survival was performed both from day 0 and day 90 of dialysis initiation. In the day 0 analysis, cumulative survival was higher for PD patients than for hemodialysis patients (HR 0.92; 95% CI 0.86–1.00; P = 0.04). The greatest benefit to starting with PD was seen in patients aged <65 years without diabetes and comorbidity. Interestingly, survival analysis at day 90 found that these differences between modality completely disappeared, suggesting that the excess mortality in these well-matched, younger, less-comorbid patients starting hemodialysis could be the result of adverse effects of hemodialysis. It is tempting to speculate that this finding reflects the potentially harmful effects of starting hemodialysis in these lower-risk patients observed by Rosansky and coauthors.2 Other potential explanations that might favor initial use of PD are improved preservation of residual renal function in PD and need for catheters for vascular access in hemodialysis and their associated risks. It must be emphasized that however good the matching, propensity-score analysis is not the same as randomization.
Relatively few studies have compared outcomes of home hemodialysis with other RRT modalities. The UK Renal Registry compared outcomes of home hemodialysis patients (n = 225, 1997–2005) with age-matched and sex-matched incident PD, hospital and satellite hemodialysis patients with follow-up until the end of 2006.8 Home hemodialysis patients had the best survival although there was a median delay of 12 months in switching to this modality after dialysis initiation. The authors acknowledged that this survival advantage may be a result of both selection and survivor bias as well as social or psychological factors and increased dialysis treatment.
Survival with short-daily hemodialysis and the impact of dialysis duration, frequency, dose and site (home [n = 189] versus hospital [n = 73]), was investigated in a European and US multicenter study.9 Four factors were independently associated with survival: age (HR 1.05), secondary renal disease (HR 2.30), weekly dialysis hours (HR 0.84), and home dialysis (HR 0.50). The benefits of short-daily hemodialysis are supported by the results of the randomized controlled Frequent Hemodialysis Network (FHN) Daily Trial showing that more frequent hemodialysis resulted in improved health status and reduced left ventricular hypertrophy.10
In conclusion, recent studies indicate that starting dialysis early solely based on an eGFR of 7–15 ml/min/1.73 m2 is not justified and could in fact be harmful. Although other studies indicate that starting below 7 ml/min/1.73 m2 is not necessarily associated with increased mortality risk, these results may reflect survivor bias. The whole idea of dependency on this single metric may in fact be flawed; alternative systematic measures of the clinical requirement to commence dialysis are required. As the commencement of dialysis is likely itself a risk, perhaps more so with hemodialysis than with PD, patients should not be exposed to dialysis until clinically necessary. With respect to modality choice, the marginal differences in outcome between PD and in-center hemodialysis no longer justify the lack of access to PD, which for many patients is the ideal home-based treatment. For selected patients, home hemodialysis (with its option to increase dialysis hours) and more-frequent hemodialysis,10 are promising.
Key Advances
Starting dialysis early, solely on the basis of an estimate of renal function, does not necessarily confer benefit2, 3, 4, 5
Estimated glomerular filtration rate may not be the right tool to determine when to start dialysis2
Alternative approaches, including more-frequent hemodialysis10 and home-based therapies (peritoneal dialysis6, 7 and home hemodialysis8, 9), are promising
Competing interests statement
The authors declare competing interests.
Top of pageReferences
1.US Renal Data System. USRDS 2009 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States (National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, USA, 2009).
2.Rosansky, S. J., Eggers, P., Jackson, K., Glassock, R. & Clark, W. F. Early start of hemodialysis may be harmful. Arch. Intern. Med. doi: 10.1001/archinternmed.2010.415.
Article3.Cooper, B. A. et al. A randomized, controlled trial of early versus late initiation of dialysis. N. Engl. J. Med. 363, 609–619 (2010).
ArticlePubMedISI4.Hwang, S.-J. et al. Impact of the clinical conditions at dialysis initiation on mortality in incident haemodialysis patients: a national cohort study in Taiwan. Nephrol. Dial. Transplant. 25, 2616–2624 (2010).
ArticlePubMedISI5.Evans, M. G., Tettamanti, G., Nyrén, O., Bellocco, R., Fored, C. M. & Elinder, C.-G. No survival benefit from early-start dialysis in a population-based, inception cohort study of Swedish patients with chronic kidney disease. J. Intern. Med. doi: 10.1111/j.1365-2796.2010.02280.x.
Article6.Mehrotra, R., Chiu, Y.-W., Kalantar-Zadeh, K., Bargman, J. & Vonesh, E. Similar outcomes with hemodialysis and peritoneal dialysis in patients with end-stage renal disease. Arch. Intern. Med. doi: 10.1001/archinternmed.2010.352.
Article7.Weinhandl, E. D., Foley, R. N., Gilbertson, D. T., Arneson, T. J., Snyder, J. J. & Collins, A. J. Propensity-matched mortality comparison of incident hemodialysis and peritoneal dialysis patients. J. Am. Soc. Nephrol. 21, 499–506 (2010).
ArticlePubMedISI8.Nitsch, D. et al. Outcomes in patients on home haemodialysis in England and Wales, 1997–2005: a comparative cohort analysis. Nephrol. Dial. Transplant. doi: 10.1093/ndt/gfq561.
Article9.Kjellstrand, C. et al. Survival with short-daily hemodialysis: association of time, site, and dose of dialysis. Hemodial. Int. 14, 464–470 (2010).
ArticlePubMedISI10.The FHN Trial Group. In-center hemodialysis six times per week versus three times per week. N. Engl. J. Med. doi: 10.1056/NEJMoa1001593.
ArticlePubMedISITop of pageAuthor affiliations
B. Lindholm & S. Davies
Divisions of Baxter Novum and Renal Medicine, Department of Clinical Science, Intervention and Technology, Karolinska Institutet, K56 Karolinska University Hospital Huddinge, S-141 86 Stockholm, Sweden (B. Lindholm). Department of Nephrology, Keele University, University Hospital of North Staffordshire, Royal Infirmary, Princess Road, Stoke-on-Trent, Staffordshire ST4 7LN, UK (S. Davies).
Correspondence to: B. Lindholm bengt.lindholm@ki.se
Published online 28 January 2011
http://www.nature.com/nrneph/journal/v7/n2/full/nrneph.2010.178.html
I Hate Dialysis Message Board
Author
