The Dialysis Outcomes and Practice Patterns Study (DOPPS) identified 6 guideline-driven, modifiable haemodialysis practices associated with increased mortality risk  : catheter access (as opposed to arteriovenous fistula), Kt/V < 1.2, serum phosphorus (PO4) > 5.5 mg/dL, serum albumin < 3.5 g/dL, interdialytic weight gain (IDWG) > 5.7%, and haemoglobin (Hb) < 11 g/dL. Associated relative mortality risk ranged from 1.11 (PO4 > 5.5 mg/dL) to 1.38 (albumin < 3.5 g/dL).
We examined these 6 modifiable risk factors using baseline data from the aggregate sample and the Italian subsample of the MONITOR-CKD5 study, and compared them to the published 2002-03 US DOPPS II results.
MONITOR-CKD5 is a prospective 24-month pharmacoepidemiological study of haemodialysis patients with renal anaemia started on Binocrit® in 10 European countries which examines the multi-level determinants, predictors and clinical outcomes of biosimilar epoetin alfa (Binocrit®/Sandoz) in haemodialysis patients.2 
This interim analysis evaluates 2087 patients enrolled from 112 centres in 10 European countries who completed 26,552 monthly visits at the time of this interim analysis. Patients’ baseline data are classified according to the DOPPS definitions of the 6 modifiable haemodialysis practices associated with increased mortality risk. The MONITOR-CKD5 data from the current study are compared to the 2002-2003 DOPPS II study.
Sample size by country for this interim analysis is:
- Austria: n = 140
- France: n = 64
- Germany: n = 867
- Italy: n = 312
- Poland: n = 120
- Romania: n = 381
- Slovenia: n = 18
- Spain: n = 115
- Switzerland: n = 25
- United Kingdom: n = 45
The mean age of patients in Italy was 67.8 ± 14.4 years, and the majority (62%) of patients were male. Age distribution by gender is shown in Figure 1. The mean time on dialysis was 3.7 ± 5.2 years. The majority of patients (85%) were already receiving ESA therapy at study start.
The most common primary aetiologies of chronic renal failure were diabetic nephropathy (20%), chronic glomerulonephritis (19%), renal vascular disease (19%), and tubular interstitial nephropathy (13%); in 20%, aetiology was unknown.
The most prevalent co-morbidities were hypertension (83%), type 2 diabetes (29%), and coronary disease (25%).
Table 1 (Figure4) shows the 6 modifiable risk factors for the MONITOR-CKD5 aggregate and Italian data versus the 2002-2003 US DOPPS II sample (* denotes region with superior result).
Prevalence of 3 of the modifiable risk factors was significantly lower in the Italian MONITOR-CKD5 sample than in the 2002-03 DOPPS II US sample including use of catheters, serum phosphate > 5.5 mg/dL, and interdialytic weight gain > 5.7%. Rates of Kt/V < 1.2 and serum albumin < 3.5 g/dL were significantly higher in Italy.
Rates of anemia were higher in the MONITOR-CKD5 study (both Italian and aggregate samples), however, when adjusted for updated guideline recommendations for Hb levels these rates decreased to half (in the aggregate sample) of the anemia rate in the previous decade in the DOPPS II sample and to three-quarters in the Italian sample.
Differences in risk factor were seen between Italy and the aggregate MONITOR-CKD5 sample and were further explored. In fact, variation between countries was significant for all of the risk factors except interdialytic weight gain (see Figure 2).
Figure 3 shows the Italian MONITOR-CKD5 physicians’ general impression of the importance of the 6 modifiable factors.
The management of the DOPPS-identified, guideline-driven modifiable risk factors in haemodialysis in 2010-2012 in Europe in general, and in Italy specifically, has come to exceed known 2002-2003 levels for the US for most of the modifiable risk factors; yet, there is room for gain.
It is not surprising that haemoglobin levels in 2010-2012 are significantly lower than 2002-2003 levels given the reduction in target haemoglobin in ESA labels and best practice guidelines. Yet, country disparities in haemoglobin achievement, as well as in other risk factors, persist.
Physicians’ impressions of the rank importance of the risk factors are not consistent with the evidence regarding added risk of the modifiable factors to mortality.