• The Amgen-Fresenius research initiative: first publications on a large EU dialysis patient cohort (ARO database)
• Summary of Published Results
• v2.0 IHQ release 02 March 2011
• ARO
• The ARO research initiative
• ARO-2: new cohort
• Publications
• An Epidemiological Study of Hemodialysis Patients Based on the European Fresenius Medical Care Hemodialysis Network: Results of the ARO Study
• Angel de Francisco, Joseph Kim, Stefan Anker, Vasily Belozeroff, Bernard Canaud, Charles Chazot, Tilman Drüeke, Kai-Uwe Eckardt, Jürgen Floege, Florian Kronenberg, Iain Macdougall, Daniele Marcelli, Bart Molemans, Jutta Passlick-Deetjen, Guntram Schernthaner, Peter Stenvinkel, David Wheeler, Bruno Fouqueray and Pedro Aljama
• de Francisco et al Nephron Clinical Practice 2011;118:c143-c154
• Objectives
• Study population
• Open-cohort study design
• Baseline demographics
• Patient numbers by country
• CKD aetiology (N=8963)
• Dialysis parameters during first 6 months of follow up
• Achieved Hb during the first 6 months of follow-up by dialysis vintage
• Laboratory parameters during first 6 months of follow up
• Medication use during first 6 months of follow up
• Patient outcomes during entire study period
• Causes of deaths during the entire study period (n=1678)
• Discussion points raised in analysis
• The diagnosis of CKD aetiology was inconsistent across countries
• Mortality rates /1000 pt years ranged from 73-182 (avg 124) vs. 140 (EuroDOPPS), 220 (US DOPPS) and 210 (USRDS)
• Lower diabetes rates in ARO (25%) vs. DOPPS (36%) and USRDS (55%)
• Incident patients had higher mortality and ESA use than prevalent patients
• Study limitations
• No data from UK or Germany
• Forty-four centres excluded as a result of inadequate data capture on key dialysis parameters
• Data from a single dialysis provider (FMC)
• Summary
• 8963 patients were included in the analysis giving a total of 12,194 patient-years of follow-up.
• Overall, patient mortality is higher in incident patients than prevalent patients.
• Patient characteristics and treatment patterns vary widely among the countries in the study owing to varied:
– Distribution of comorbid conditions
– Availability of health care resources
– Physician training
• Hemoglobin Variability Does Not Predict Mortality in European Hemodialysis Patients
• Kai-Uwe Eckardt, Joseph Kim, Florian Kronenberg, Pedro Aljama, Stefan D Anker, Bernard Canaud, Bart Molemans, Peter Stenvinkel, Guntram Schernthaner, Elizabeth Ireland, Bruno Fouqueray and Iain C Macdougall
• Eckardt et al J Am Soc Nephrol 21: 1765–1775, 2010
• Background and rationale
• Significant Hb variability in US HD populations
• Unclear whether associated with mortality
• So far, no data in European dialysis cohorts
• Objectives
• Describe and characterize Hb variability in a European population of HD patients
• Identify predictors of high Hb variability
• Evaluate the association between Hb variability and all-cause and cardio-vascular mortality
• Study population
• Methods
• Exposure period: 6 months
• Observation period: up to 18 months, immediately following exposure
• Measures of Hb variability
– Standard deviation (SD)
– Residual SD
– Time spent in target (11-12.5 g/dL)
– Fluctuations across thresholds (CL, CT, CH, LAL, LAH, HA)
– Area under the Curve (AUC)
• Predictors of high variability were evaluated using logistic regression
• Cox regression was used to examine the association between Hb variability and mortality
• Methods
Area Under the Curve (AUC)
Area Under the Curve (AUC)
• To capture both deviations of Hb from the mean and frequencies of fluctuations across time in a single quantitative index
• AUC was calculated for each individual patient’s 6-month Hb profile
• Incident patients had greater Hb variability and spent less time in target
• Categories of Hb fluctuation across thresholds
• Incident patients more likely to be in consistently low or high-amplitude Hb categories
• Predictors of Hb variability
• Hb variability expressed using SD, residual SD, time in target, and AUC is not associated with an increased risk of mortality after multivariable adjustment
• Categories of Hb fluctuation across thresholds
• Patients with consistently low Hb levels have the highest risk of mortality
• Study limitations
• Observational study:
– difficult to determine cause-and-effect relationships
– clinical database was not originally established as a research tool
– database is from a single private dialysis provider (FMC)
– generalizability to patients at other providers and countries without FMC centers remains unclear
• 44% of patients excluded from analysis because of open cohort design (52% of these were incident); may have resulted in selection bias.
• Absence of information on iron therapy which may influence Hb variability or survival.
• Median observation period ~1 year; precludes conclusions on longer-term mortality.
• Although N>5000, numbers of events in subgroups were relatively small.
• DOPPS: Risk of mortality is higher in facilities with higher Hb variability
• Discussion of findings on Hb variability in DOPPS and Medicare analyses
• Inter-facility Hb variability:1
– seen to be reduced between 1996-2008 in most countries assessed
– mainly dependent on practice patterns
– may be proposed as an indicator of care
• Inter-patient ESA dose increases and younger age were linked with raised mortality risk and increased Hb variability1
• Intra-patient Hb variability association with mortality risk was weak, and inconsistent after adjustment for concurrent disease severity2
• It remains to be seen whether decrease in Hb variability in an individual facility over time will translate into improved outcomes3
Ø Hb variability may indicate disease severity and difference in quality of practice patterns, but by itself is not a good indicator of risk1-3
• Summary and conclusions
• Hb variability occurs in the European HD population to a similar extent as in US HD populations
• It is related to various patient characteristics, comorbidities, and hospitalizations but is not an independent risk factor for all-cause or cardio-vascular mortality
• However, patients with consistently low Hb levels have the highest risk of mortality
• Backup
• Variables of adjustment
• Demographics
– age, gender, country, BMI, smoking status
• Medical history
– CKD aetiology, history of diabetes, history of CVD, history of Cancer
• Dialysis parameters
– Vintage, vascular access type, Kt/V, blood flow
• Markers of inflammation
– Serum albumin, CRP
• CVD medications
– Antihypertensive drugs, ACE inhibitors, oral anticoagulants, anti-aggregants
• BMM medications
– Vitamin D, phosphate binders
• Other lab parameters
– PTH, calcium, phosphate, Hb, ferritin, cholesterol, blood leucocytes
• Miscellaneous
– Hospitalization, change in vascular access type
• AUC is highly correlated with within-person SD but not with time spent in target
• Distribution of AUC is consistent with categories of fluctuation across thresholds
• Serum iPTH, Calcium and Phosphate, and the Risk of Mortality in a European Haemodialysis Population
• Jürgen Floege, Joseph Kim, Elizabeth Ireland, Charles Chazot, Tilman Drueke, Angel de Francisco, Florian Kronenberg, Daniele Marcelli, Jutta Passlick-Deetjen, Guntram Schernthaner, Bruno Fouqueray and David C. Wheeler on behalf of the ARO Investigators
• Floege J et al. NDT 2010; published online Dec 10: doi: 10.1093/ndt/gfq219
• Background and rationale
• A number of US observational studies reported an increased mortality risk with higher intact parathyroid hormone (iPTH), calcium and/or phosphate.
• The existence of such a link in a European haemodialysis population was explored as part of the Analysing Data, Recognising Excellence and Optimising Outcomes (ARO) Chronic Kidney Disease (CKD) Research Initiative.
• Objectives
• The aim of the present analysis was to examine the relation between levels of soluble markers of mineral bone disease (MBD: iPTH, total serum calcium and serum phosphate) and long-term mortality in 7970 hemodialysis patients treated at Fresenius Medical Care facilities in Europe.
• Study population
• Methods
• Measurements of iPTH, calcium and phosphate were averaged over first quarter of follow-up, then divided into categories
• All-cause mortality was the primary outcome of interest
• Crude and adjusted hazard ratios (HR) for mortality were determined using baseline (i.e. fixed-covariate) Cox regression models
• Time-dependent Cox analysis was performed to evaluate any potential effects of updating exposure and selected covariates over time
• Baseline data by iPTH level
• Baseline medical data by iPTH level
• Baseline and time-dependent Cox regression for all-cause mortality
• Adjusted risk for iPTH levels
• Adjusted risk for calcium levels
• Adjusted risk for phosphate levels
• Adjusted risk for iPTH in diabetic vs. non-diabetic patients
• Summary of the key findings
• Serum iPTH:
– In the adjusted baseline Cox analysis, the HR estimates were U-shaped. Patients with iPTH level below or above the 2003 KDOQI target range of 150 to 300 pg/mL had a greater risk of death.
– The adjusted time-dependent analysis confirmed the U-shaped pattern.
• Serum calcium:
– The adjusted baseline analysis showed a higher risk of death in patients with high serum calcium levels (> 2.75 mmol/L) compared to those within the target range.
– The adjusted time-dependent analysis showed that both patients with low calcium level and high calcium level had increased risk of death.
• Serum phosphate:
– The adjusted baseline analysis showed a U-shaped pattern similar to that for iPTH. Both low and high serum phosphate increased risk of death.
– The adjusted time-dependent analysis showed increased risk of death for low phosphate only.
• Comparative data from published studies on MBD and mortality in large HD populations
• CORES study shows similar U-shaped association of mortality with PTH levels in Latin America
• Study limitations
• Analyses were based on observational data and no causal inference can be made.
• Missing data were common among all the MBD markers considered, as were some potentially important confounding factors such as dialysate calcium concentration and intravenous administration of active vitamin D therapy derivatives
• Results were not stratified by incident vs prevalent patients due to the small number of incident patients.
• Adjustment for serum 25(OH) vitamin D was not made because this information was not captured in the ARO database.
• Conclusions
• Patients with iPTH, calcium and phosphate levels within the KDOQI recommended targets experienced the lowest risk of mortality compared with those outside the respective target ranges
• The data are consistent with the KDIGO recommendations on CKD-MBD target parameters, although some patients could be at increased risk of mortality compared to those treated within the KDOQI target range
• Very low or high values of iPTH and phosphate, as well as high values of calcium, should be avoided
• Backup
• Variables of adjustment
• Demographics
– age, gender, country, BMI, smoking status
• Medical history
– CKD aetiology, history of diabetes, history of CVD, history of Cancer
• Dialysis parameters
– Vintage, vascular access type, Kt/V, blood flow
• Markers of inflammation
– Serum albumin, CRP
• CVD medications
– Antihypertensive drugs, ACE inhibitors, oral anticoagulants, anti-aggregants
• BMM medications
– Vitamin D, phosphate binders
• Other lab parameters
– PTH, calcium, phosphate, Hb, ferritin, cholesterol, blood leucocytes
• Miscellaneous
– Hospitalization, change in vascular access type
