Transplant Association Registry Research Articles

Analysis of data from the ERA-EDTA Registry indicates that conventional treatments for chronic kidney disease do not reduce the need for renal replacement therapy in autosomal dominant polycystic kidney disease.

Autosomal dominant polycystic kidney disease (ADPKD) is a major cause of end-stage kidney failure, but is often identified early and therefore amenable to timely treatment. Interventions known to postpone the need for renal replacement therapy (RRT) in non-ADPKD patients have also been tested in ADPKD patients, but with inconclusive results. To help resolve this we determined changes in RRT incidence rates as an indicator for increasing effective renoprotection over time in ADPKD. We analyzed data from the European Renal Association-European Dialyses and Transplant Association Registry on 315,444 patients starting RRT in 12 European countries between 1991 and 2010, grouped into four 5-year periods. Of them, 20,596 were due to ADPKD. Between the first and last period the mean age at onset of RRT increased from 56.6 to 58.0 years. The age- and gender-adjusted incidence rate of RRT for ADPKD increased slightly over the four periods from 7.6 to 8.3 per million population. No change over time was found in the incidence of RRT for ADPKD up to age 50, whereas in recent time periods the incidence in patients above the age of 70 clearly increased. Among countries there was a significant positive association between RRT take-on rates for non-ADPKD kidney disease and ADPKD. Thus, the increased age at onset of RRT is most likely due to an increased access for elderly ADPKD patients or lower competing risk prior to the start of RRT rather than the consequence of effective emerging renoprotective treatments for ADPKD.

Renal unit characteristics and patient education practices that predict a high prevalence of home-based dialysis in Australia.

The proportion of patients using home dialysis in Australia varies from 6% to 62% between renal units. The aim of this study was to determine if the variance is attributed to any underlying renal unit factors including pre-end stage education practices. An online survey was distributed to all Australian units that offered home dialysis. Logistic regression was performed to estimate the effects of renal unit characteristics on the binary outcome of <30% versus ≥ 30% of patients using home dialysis, and for ≥ 10% of patients using home haemodialysis (HHD) dialysis specifically. Prevalent home dialysis rates were sourced from the Australia and New Zealand Dialysis and Transplant Association registry. 33 of 43 units (77%) completed the survey. Factors shown to predict ≥ 30% of patients using home dialysis were; a metropolitan based renal unit compared with a rural or remote unit (OR 1.08, 95% CI 1.01-1.15), a New South Wales unit compared with other states (OR 1.13, 95% CI 1.04-1.22), and a unit that offered multiple group education sessions per year (OR 1.01, 95% CI 1.01-1.02). A unit that offered >1 h of pre-end stage education per patient, compared with ≤ 1 h predicted more than 10% of patients on HHD (OR 2.84, 95% CI 1.17-6.90). Our data suggest certain pre-end stage education practices are significantly associated with home dialysis rates above the national average. The consistent above average home dialysis rates witnessed in New South Wales appear to be the result of renal unit culture, education strategies and policies that support 'home dialysis first'.

From sugar to kidney: A never-ending battle.

From sugar to kidney: A never-ending battle.

Likelihood of children with end-stage kidney disease in Europe to live with a functioning kidney transplant is mainly explained by nonmedical factors

Registry data can be used to assess associations between medical and health-policy factors and the likelihood of children on renal replacement therapy (RRT) to live with a functioning kidney transplant in Europe. A survey questionnaire was distributed among renal registry representatives in 38 European countries, and additional data was obtained from the European Society for Pediatric Nephrology/European Renal Association-European Dialysis and Transplant Association (ESPN/ERA-EDTA) registry. Thirty-two countries with a pediatric RRT program responded. The median percentage of children by country on RRT with a functioning transplant was 62 % (interquartile range 39-77). One per million population increase in donation rate from deceased donors was associated with a 5 % increase in the percentage of functioning transplants; the existence of an intermediate and high pediatric priority policy doubled and tripled this percentage, respectively, compared with no priority, whereas an increase in living donor pediatric kidney transplant rate of one per million children was associated with a 14 % higher percentage of functioning transplants. The percentage of functioning transplants was also strongly associated with the gross domestic product (GDP). Considerable variations exist in the percentages of prevalent pediatric RRT populations with functioning renal transplants across Europe. A macroeconomic indicator such as GDP is the most important determinant of these international differences. Efforts should be made for living donation programs and pediatric allocation priority to increase access to kidney transplantation for children.

Underweight, overweight and obesity in paediatric dialysis and renal transplant patients

The prevalence of childhood overweight is rising worldwide, but in children on renal replacement therapy (RRT) a poor nutritional status is still the primary concern. We aimed to study the prevalence of, and factors associated with, underweight and overweight/obesity in the European paediatric RRT population. Moreover, we assessed the evolution of body mass index (BMI) after the start of RRT. We included 4474 patients younger than 16 years from 25 countries of whom BMI data, obtained between 1995 and 2010, were available within the European Society for Paediatric Nephrology/European Renal Association-European Dialysis and Transplant Association Registry. Prevalence estimates for under- and overweight/obesity were calculated using age and sex-specific criteria of the World Health Organization (WHO, 0-1 year olds) and the International Obesity Task Force cut-offs (2-15 year olds). The prevalence of underweight was 3.5%, whereas 20.8% of the patients were overweight and 12.5% obese. Factors associated with being underweight were receiving dialysis treatment and infant age. Among transplanted recipients, a very short stature (OR: 1.64, 95% CI: 1.40-1.92) and glucocorticoid treatment (OR: 1.23, 95% CI: 1.03-1.47) were associated with a higher risk of being overweight/obese. BMI increased post-transplant, and a lower BMI and a higher age at the start of RRT were associated with greater BMI changes during RRT treatment. Overweight and obesity, rather than underweight, are highly prevalent in European children on RRT. Short stature among graft recipients had a strong association with overweight, while underweight appears to be only a problem in infants. Our findings suggest that nutritional management in children receiving RRT should focus as much on the prevention and treatment of overweight as on preventing malnutrition.

Survival in dialysis patients is different between patients with diabetes as primary renal disease and patients with diabetes as a co-morbid condition

A previous study in Dutch dialysis patients showed no survival difference between patients with diabetes as primary renal disease and those with diabetes as a co-morbid condition. As this was not in line with our hypothesis, we aimed to verify these results in a larger international cohort of dialysis patients. For the present prospective study, we used data from the European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Registry. Incident dialysis patients with data on co-morbidities (n = 15,419) were monitored until kidney transplantation, death or end of the study period (5 years). Cox regression was performed to compare survival for patients with diabetes as primary renal disease, patients with diabetes as a co-morbid condition and non-diabetic patients. Of the study population, 3,624 patients (24%) had diabetes as primary renal disease and 1,193 (11%) had diabetes as a co-morbid condition whereas the majority had no diabetes (n = 10,602). During follow-up, 7,584 (49%) patients died. In both groups of diabetic patients mortality was higher compared with the non-diabetic patients. Mortality was higher in patients with diabetes as primary renal disease than in patients with diabetes as a co-morbid condition, adjusted for age, sex, country and malignancy (HR 1.20, 95% CI 1.10, 1.30). An analysis stratified by dialysis modality yielded similar results. Overall mortality was significantly higher in patients with diabetes as primary renal disease compared with those with diabetes as a co-morbid condition. This suggests that survival in diabetic dialysis patients is affected by the extent to which diabetes has induced organ damage.

Growth and maturation improvement in children on renal replacement therapy over the past 20 years

The attainment of normal growth and maturation remains a major challenge in the management of children and adolescents requiring renal replacement therapy (RRT). We compared growth and maturation in 384 German children with RRT who were followed between 1998 and 2009 with 732 children who were enrolled in the European Dialysis and Transplant Association (EDTA) Registry from 1985 to 1988; of these children, 78 and 88 %, respectively, were transplanted. The data on the German patients included in the EDTA registry did not differ significantly from those of the patients from other European countries. Overall, the mean height standard deviation score (SDS) has improved over the past 20 years from -3.03 to -1.80 (p < 0.001). Until the age of 6 years, the difference in height SDS was not significant, whereas it improved significantly in adolescence (-3.40 vs. -1.52; p < 0.001). Significant improvements in the delay of the pubertal growth spurt, age at menarche, bone maturation and body mass index (BMI) were noted in the recent German group compared to the EDTA group (each p < 0.001). Our findings demonstrate a marked improvement of growth and maturation in paediatric patients on RRT during the past 20 years.

Renal replacement therapy in Europe-a summary of the 2010 ERA-EDTA Registry Annual Report.

BackgroundThis study provides a summary of the 2010 European Renal Association–European Dialysis and Transplant Association (ERA–EDTA) Registry Annual Report (available at www.era-edta-reg.org).MethodsThis report includes data on renal replacement therapy (RRT) using data from the national and regional renal registries in 29 countries in Europe and bordering the Mediterranean Sea. Individual patient data were received from 27 registries, whereas 18 registries contributed data in aggregated form. We present incidence and prevalence of RRT, transplant rates, survival probabilities and expected remaining lifetimes. The latter two are solely based on individual patient records.ResultsIn 2010, the overall incidence rate of RRT for end-stage renal disease (ESRD) among all registries reporting to the ERA–EDTA Registry was 123 per million population (pmp) (n = 91 798). The highest incidence rate was reported by Turkey (252pmp) and the lowest reported by Montenegro (21 pmp). The overall prevalence of RRT for ESRD at 31 December 2010 among all registries reporting to the ERA–EDTA Registry was 741 pmp (n = 551 005). The prevalence varied from 124 pmp in Ukraine to 1580 pmp in Portugal. The overall number of renal transplantations performed in 2010 among all registries was 29.2 pmp (n = 21 740). The highest overall transplant rate was reported from Spain, Cantabria (73 pmp), whereas the highest transplant rate for living donor kidneys was reported from the Netherlands (28 pmp). For patients who started RRT between 2001 and 2005, the unadjusted 5-year patient survival on RRT was 46.2% [95% confidence interval (CI) 46.0–46.3], and on dialysis 38.6% (95% CI 38.5–38.8). The unadjusted 5-year patient survival after the first renal transplantation performed between 2001 and 2005 was 86.6% (95% CI 86.1–87.1) for deceased donor kidneys and 94.1% (95% CI 93.4–94.8) for living donor kidneys.

Impact of graft loss among kidney diseases with a high risk of post-transplant recurrence in the paediatric population

Some kidney diseases tend to recur in the renal allograft after transplantation. We studied the risk of graft loss among primary renal diseases known for their high risk of recurrence and compared it with that of patients with hypoplasia and/or dysplasia. Within the European Society of Paediatric Nephrology and European Renal Association and European Dialysis and Transplant Association (ESPN/ERA-EDTA) registry, we studied children from 33 countries who received a kidney transplant before the age of 20 between 1990 and 2009. Patients were censored after 5 years of follow-up and cumulative incidence competing risk analysis was used to calculate survival curves. Patients with focal and segmental glomerulosclerosis (FSGS), haemolytic uraemic syndrome (HUS), membranoproliferative glomerulonephritis Type I or II (MPGN), IgA nephropathy or Henoch Schönlein Purpura (HSP/IgA) or systemic lupus erythomatosus (SLE) underwent pre-emptive transplantation significantly less often than patients with hypoplasia and/or dysplasia. The rate of living donation was lower among patients with FSGS and SLE than in patients with hypoplasia and/or dysplasia. In comparison with hypoplasia and/or dysplasia patients with a risk of 14.4%, the 5-year risk of graft loss was significantly increased in patients with FSGS (25.7%) and MPGN (32.4%) while it was not significantly increased in children with HUS (18.9%), HSP/IgA (16.3%) or SLE (20.3%). One-year graft survival strongly improved among HUS patients from 17.1% in 1995-1999 to 3.6% in 2005-2009 and was not accompanied by a decrease in the number of transplantations. The risk of graft loss is increased among specific causes of renal failure with a high risk of post-transplant recurrence. It seems likely that, due to anticipation of such risk, physicians perform less pre-emptive transplantation and provide fewer grafts from living related donors in patients with these conditions. Improved risk stratification by physicians, resulting in the identification of patients with HUS at higher or lower risk of recurrence, might explain the much improved graft survival rates.

New primary renal diagnosis codes for the ERA-EDTA

The European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Registry has produced a new set of primary renal diagnosis (PRD) codes that are intended for use by affiliated registries. It is designed specifically for use in renal centres and registries but is aligned with international coding standards supported by the WHO (International Classification of Diseases) and the International Health Terminology Standards Development Organization (SNOMED Clinical Terms). It is available as supplementary material to this paper and free on the internet for non-commercial, clinical, quality improvement and research use, and by agreement with the ERA-EDTA Registry for use by commercial organizations. Conversion between the old and the new PRD codes is possible. The new codes are very flexible and will be actively managed to keep them up-to-date and to ensure that renal medicine can remain at the forefront of the electronic revolution in medicine, epidemiology research and the use of decision support systems to improve the care of patients.

Timing and Outcome of Renal Replacement Therapy in Patients with Congenital Malformations of the Kidney and Urinary Tract

Congenital anomalies of the kidney and urinary tract (CAKUT) are the leading cause of ESRD in children, but the proportion of patients with individual CAKUT entities progressing to ESRD during adulthood and their long-term clinical outcomes are unknown. This study assessed the age at onset of renal replacement therapy (RRT) and patient and renal graft survival in patients with CAKUT across the entire age range. Patients with CAKUT were compared with age-matched patients who were undergoing RRT for other renal disorders on the basis of data from the European Renal Association-European Dialysis and Transplant Association Registry. Competing risk and Cox regression analyses were conducted. Of 212,930 patients commencing RRT from 1990 to 2009, 4765 (2.2%) had renal diagnoses consistent with CAKUT. The proportion of incident RRT patients with CAKUT decreased from infancy to childhood and then increased until age 15-19 years, followed by a gradual decline throughout adulthood. Median age at RRT start was 31 years in the CAKUT cohort and 61 years in the non-CAKUT cohort (P<0.001). RRT was started earlier (median, 16 years) in patients with isolated renal dysplasia than in those with renal hypoplasia and associated urinary tract disorders (median, 29.5-39.5 years). Patients with CAKUT survived longer than age- and sex-matched non-CAKUT controls because of lower cardiovascular mortality (10-year survival rate, 76.4% versus 70.7%; P<0.001). CAKUT leads to ESRD more often at adult than pediatric age. Treatment outcomes differ from those of acquired kidney diseases and vary within CAKUT subcategories.

Prevalence and predictors of the sub-target Hb level in children on dialysis

Anaemia is a common and potentially treatable co-morbidity of end-stage renal disease. We aimed to determine the prevalence of the sub-target haemoglobin (Hb) level among European children on dialysis and to identify factors associated with a low Hb level. From the European Society for Paediatric Nephrology (ESPN)/European Renal Association-European Dialysis Transplant Association (ERA-EDTA) registry, data were available on 2351 children between 1 month and 18 years of age, totalling 5546 measurements from 19 countries. The mean Hb level was 10.8 g/dL (5th-95th percentiles, 7.4-13.9). Among those above 2 years of age, the mean Hb level was 10.9 g/dL (11.4% below 8.5 g/dL), while it was 10.3 g/dL among those below 2 years (11.2% below 8.0 g/dL). A total of 91.2% of the patients were on an erythropoiesis-stimulating agent (ESA). Hb levels increased with age and were higher in peritoneal dialysis compared with haemodialysis patients. Patients with congenital anomalies of the kidney and urinary tract showed the highest Hb levels, and those with cystic kidney diseases or metabolic disorders the lowest ones. Ferritin levels between 25 and 50 ng/mL were associated with the highest Hb levels. We found a weak inverse association between parathyroid hormone (PTH) and Hb. Whereas standardized blood pressure (BP) was not elevated in patients with above-target Hb, elevated systolic BP z-score was noted in those with sub-target Hb levels. Sub-target Hb levels remain common in children on dialysis, in spite of virtually all children being treated with ESA; although we cannot exclude under-dosing. Optimal ferritin levels seemed to be slightly lower in children (25-50 ng/mL) than those in adults. Other risk factors for sub-target Hb are dialysis modality and a high PTH level.

Are the high mortality rates in dialysis patients mainly due to cardiovascular causes?

It is a well-known fact that chronic dialysis patients have a markedly increased mortality compared to the general population. In a recently published analysis of the European Renal Association–European Dialysis and Transplant Association Registry, the mortality rate in incident dialysis patients was 192 per 1000 person-years, while it was only 12.05 in the general population [1]. Despite the emphasis on the importance of accelerated atherosclerosis and cardiovascular death in dialysis patients [2], the excess mortality in them was due to both cardiovascular (39%) and noncardiovascular (51%) causes. The distribution of the causes of death was not different from that in the general population. The standardized mortality ratio for cardiovascular death was 42.9 in dialysis patients compared to 4.9 in the general population. For non-cardiovascular death, these figures were 57.1 and 7.0. This ‘normal’ distribution of mortality causes may lead to a different view on risk factors for mortality in dialysis patients. Also, factors that have been linked to cardiovascular death are often associated with non-cardiovascular mortality. This has been found for C-reactive protein [3], fetuin A [4] and possibly troponin T [5], making it likely that death from various causes is often associated with the presence of an acute-phase reaction.

Prevalence and predictors of anaemia in european children on dialysis

Despite recommendations for the use of iron therapy and erythropoiesis stimulating agents (ESA) to achieve optimal haemoglobin (Hb) levels, anaemia is a significant problem in children with Chronic Kidney Disease...

A Renal Transplantation and Hemodialysis Cost-Utility Analysis in Patients with End-Stage Renal Disease

Background: The purpose of this study was to examine the cost-effectiveness of renal transplantation and hemodialysis among end-stage renal disease patients.Methods: Empirical data on treatment costs were collected from five hospitals in Korea. We used European Renal Association-European Dialysis and Transplant Association registry data for transition probability. Quality adjusted life year (QALY) values were derived from the literature. A Markov model was used for predicting the cost utility of transplantation and hemodialysis over a 10-year period.Results: Renal transplantation was less costly and resulted in a better outcome than hemodialysis. The cost per QALY gained was 19,450 thousand won in transplantation patients, whereas it was 36,514 thousand won per QALY gained in hemodialysis patients. Conclusions: Although the cost of the first year after transplantation was expensive, transplantation was more effective over 2 years and was less costly than hemodialysis. The results suggest that transplantation is more cost-effective than hemodialysis in Korea.Key Words: Cost-effectiveness analysis, Cost-utility analysis, Renal transplantation, Hemodialysis, End-stage renal disease (ESRD)중심 단어: 비용-효과분석, 비용-효용분석, 신장이식, 혈액투석, 말기신부전책임저자：함명일, 아산시 신창면 읍내리 646순천향대학교 의료과학대학 보건행정경영학과, 336-745Tel: 041-530-3035, Fax: 041-530-3085E-mail: hmi@sch.ac.kr접수일 : 2010년 4월 5일, 심사일 : 2010년 5월 28일게재승인일 : 2010년 7월 6일

Level of renal function in patients starting dialysis: an ERA-EDTA Registry study

The aims of this European study were (i) to compare the level of renal function at the start of dialysis between age groups, gender, primary renal disease, comorbid conditions, treatment modality, time periods and countries, and (ii) to determine which baseline characteristics are associated with the level of renal function at the start of dialysis. Renal registries participating in the European Renal Association-European Dialysis and Transplant Association Registry provided data on serum creatinine 0-4 weeks before the start of dialysis in incident dialysis patients in 1999 and 2003. Data were available in 11 472 patients from nine renal registries. Glomerular filtration rate (GFR) was estimated by the four-variable Modification of Diet in Renal Disease equation. The unadjusted median eGFR at the start of dialysis was 7.0 mL/min/1.73 m(2) in the 1999 data (median serum creatinine 7.5 mg/dL) and 7.7 mL/min/1.73 m(2) in the 2003 data (serum creatinine 7.0 mg/dL). Using linear regression with adjustment for the other covariates, older patients, males, patients with diabetes mellitus, hypertension/renal vascular disease (HT/RVD) as primary renal disease (vs glomerulonephritis), ischaemic heart disease or peripheral vascular disease and patients starting on peritoneal dialysis (PD) initiated dialysis at higher levels of eGFR (range Δ eGFR: 0.1-1.2 mL/min/1.73 m(2)). Using the same analyses, eGFR differed between countries (range: 6.5-8.6 mL/min/1.73 m(2)). During 2003, patients started dialysis at somewhat higher eGFR levels than those starting during 1999. There were also international differences in eGFR. Such differences may, at least in part, be explained by differences in creatinine measurement methods between countries and time periods. Finally, older patients, males, patients with HT/RVD or comorbidity and those starting on PD had slightly higher eGFR levels than younger patients, females, those with glomerulonephritis, without comorbidity and those starting on haemodialysis. Further research is needed into other, more clinically related factors affecting the decision to start dialysis.

Determinants of eGFR at start of renal replacement therapy in paediatric patients

Few studies have investigated the determinants of glomerular filtration rate (GFR) in paediatric patients starting on dialysis or with a transplant. Data were collected as part of the European Society of Paediatric Nephrology/European Renal Association-European Dialysis and Transplant Association registry from 14 European countries and referred to incident paediatric patients starting on renal replacement therapy (RRT) between 2002 and 2007 under the age of 18 years. Estimated glomerular filtration rate (eGFR) was calculated using the Schwartz formula. Data were adjusted for age, gender, treatment modality at start, primary cause of renal failure (PRD) and regions in Europe (eGFR(adj)). Median eGFR in the 938 patients starting RRT was 10.4 mL/min/1.73 m(2) (5th and 95th percentile: 4.0-26.9). Twenty-six patients (2.8%), mainly infants with Finnish-type nephropathy, started with eGFR levels >50 mL/min/1.73 m(2). Younger age, female gender, starting on dialysis and having a short time between the first visit to a paediatric nephrologist (PN) and start of RRT were associated with lower eGFR at start of RRT. Gender differences were only present during adolescent age and disappeared when using the same K value for both genders. The various PRDs showed large differences in the rate of decline in eGFR between the first visit to a PN and start of RRT; however, this did not result in differences in eGFR(adj) at start of RRT. The main determinants of eGFR at start of RRT were age, gender, treatment modality at start, and the time between the first visit to a PN and start of RRT. Research is needed to determine the consequences of these differences.

The Databases

Chronic kidney disease is now considered a public health priority, and the prevalence of this disease is approximately 10% in both North American and European countries. Such a phenomenon raises concern about the future increased incidence of ESRD. A recent analysis in the European Renal Association-European Dialysis and Transplant Association Registry shows that the incidence rates in Northern European countries have stabilized at approximately 110 per million people, a phenomenon that is associated with a parallel stabilization in the incidence of ESRD caused by diabetes. Such a stabilization has occurred in the face of an increasing prevalence of diabetes and hypertension in the general population, suggesting that this improvement may be the result of better prevention. Genetic factors, competing risks with other diseases, and other medical factors explain only in part the variability in the incidence of renal replacement therapy in European countries. Health care financing priorities have an obvious influence on the outcome of ESRD. Nonmedical factors seem to be of importance at least equal to that of medical factors. In this respect, Dialysis Outcomes and Practice Patterns Study Europe has revealed relevant differences in clinical policies that are related to ESRD treatment among European countries.

Trends in Hemodialysis Vascular Access From the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA) 2000 to 2005

Australia historically has been recognized for its high fistula use. Observational study using registry data. Adult patients registered in the Australia and New Zealand Dialysis and Transplant Association Registry on hemodialysis in Australia. Cohort year. Hemodialysis access trends were examined from 2000 to 2005 for incident patients (within 60 days of hemodialysis therapy start), patients on hemodialysis therapy for 6 to 8 months, and prevalent hemodialysis patients. Multivariate analyses were performed to examine the relationship between access type and cohort year for each group, with adjustment for age, sex, race, body mass index, late referral, smoking status, cause of end-stage renal disease, comorbidities, and dialysis vintage. During 2000 to 2005, catheter use increased from 39% to 53% in incident patients, 10% to 22% in the 6- to 8-month groups, and 6% to 13% in prevalent patients. Fistula use decreased from 56% to 43% in incident patients and 78% to 67% in the 6- to 8-month group and remained at 73% to 75% in prevalent patients. Graft use decreased in all groups. Adjustment for factors associated with access type did not significantly change these results. The registry collects only the access in use at the end of the survey period; thus, it was not possible to determine whether another access had failed or was present, but not in use. The small number of incident numbers prevented separate analysis of arteriovenous fistulas and arteriovenous grafts. Incident use of fistulas and grafts decreased, with an unexpected increase in both incident and prevalent catheters between 2000 and 2005. Adjustment for factors associated with access type did not significantly alter the trends. Changes in unidentified practice patterns, attitudes, or preferences are contributing to these trends. Ongoing evaluation of data and investigation into processes of care are required to increase functioning fistulas, together with reevaluation of the role of grafts in patients without a fistula.

Confounding effect of comorbidity in survival studies in patients on renal replacement therapy

After taking other confounding factors into account, the impact of comorbidity on mortality was investigated when comparing mortality between five European countries, dialysis modalities and renal disease groups. The study included 15 571 incident patients on renal replacement therapy (RRT) from five national or regional registries participating in the European Renal Association-European Dialysis and Transplant Association Registry that collect comorbidity data. The presence of diabetes mellitus, ischaemic heart disease, peripheral vascular disease, cerebrovascular disease and malignancy was recorded at the start of RRT. The comorbidities were each independently associated with mortality, with hazard ratios (HRs) ranging from 1.40 (95% CI: 1.30-1.51) for peripheral vascular disease to 1.65 (95% CI: 1.48-1.83) for diabetes. Age, gender, primary renal disease, modality and country together explained 14.4% of the variance in mortality; the comorbidities explained an additional 1.9%. In the comparison of renal vascular disease with glomerulonephritis, the crude HR of 2.40 (95% CI: 2.12-2.72) changed to 1.24 (95% CI: 1.09-1.41) after adjustment for age, gender, primary renal disease, treatment modality and country and to 1.06 (95% CI: 0.93-1.22) after further adjustment for the comorbidities. For the comparison between countries and other patient groups, the change in the survival estimate after adjustment for comorbidity was less. Comorbidity is an important predictor for mortality. However, after adjustment for age, gender, primary renal disease, treatment modality and country, when comparing outcomes between patient groups the influence of comorbidity may be less important than expected.