Table 2.
Citric acid–containing dialysate and mortality risk, effect of progressive adjustment
| Exposure | % of Patients | Hazard Ratio (95% Confidence Interval) | ||||
| Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | ||
| No citrate | 86 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Citrate=1 mEq/L | 2 | 1.13 (0.82 to 1.56) | 1.10 (0.81 to 1.51) | 1.09 (0.80 to 1.50) | 1.15 (0.84 to 1.56) | 1.25 (0.90 to 1.74) |
| Citrate=2 mEq/L | 9 | 1.30 (1.00 to 1.70) | 1.25 (0.96 to 1.63) | 1.24 (0.95 to 1.62) | 1.24 (0.95 to 1.62) | 1.24 (0.93 to 1.64) |
| Citrate=3+mEq/L | 3 | 1.13 (0.85 to 1.50) | 0.99 (0.76 to 1.28) | 1.00 (0.77 to 1.31) | 1.02 (0.79 to 1.31) | 0.88 (0.66 to 1.16) |
| Any citrate (versus none) | 14 (86) | 1.20 (1.02 to 1.42) | 1.12 (0.96 to 1.32) | 1.12 (0.95 to 1.32) | 1.14 (0.97 to 1.34) | 1.10 (0.92 to 1.32) |
Restricted to countries/phases (Belgium/5 and 6, Canada/6, Germany/5 and 6, Italy/5 and 6, Japan/5 and 6, Spain/6, Sweden/5 and 6, United Kingdom/5 and 6) with at least 15 patients prescribed citric acid–containing dialysate n=11,318 patients and 1862 deaths; dialysate composition assessed at a single time point: earliest of Dialysis Outcomes and Practice Patterns Study (DOPPS) study enrollment or the first study follow-up round in which citric acid–containing dialysate became available in DOPPS phase 5 (2013).
Model 1: stratified by country and phase; accounting for facility clustering.
Model 2: additionally adjusted for age and sex.
Model 3: additionally adjusted for diabetes and body mass index <18 kg/m2.
Model 4: additionally adjusted for hemodiafiltration use, catheter use, single pool Kt/V, and dialysis session length.
Model 5: additionally adjusted for total dialysate base concentration (sum of bicarbonate + acetate or citrate), dialysate calcium ≤2.5 versus >2.5 mEq/L, albumin, phosphate, calcium, and parathyroid hormone concentration (PTH); model 5 considered as potentially overadjusted for variables in the causal pathway.