Initiating haemodialysis twice-weekly as part of an incremental programme may protect residual kidney function

Abstract

Background: Initiating twice-weekly haemodialysis (2×HD) in patients who retain significant residual kidney function (RKF) may have benefits. We aimed to determine differences between patients initiated on twice- and thrice-weekly regimes, with respect to loss of kidney function, survival and other safety parameters. Methods: We conducted a single-centre retrospective study of patients initiating dialysis with a residual urea clearance (KRU) of ≥3 mL/min, over a 20-year period. Patients who had 2×HD for ≥3 months during the 12 months following initiation of 2×HD were identified for comparison with those dialysed thrice-weekly (3×HD). Results: The 2×HD group consisted of 154 patients, and the 3×HD group 411 patients. The 2×HD patients were younger (59 ± 15 versus 62 ± 15 years: P = 0.014) and weighed less (70 ± 16 versus 80 ± 18 kg: P < 0.001). More were females (34% versus 27%: P = 0.004). Fewer had diabetes (25% versus 34%: P = 0.04) and peripheral vascular disease (PVD) (13% versus 23%: P = 0.008). Baseline KRU was similar in both groups (5.3 ± 2.4 for 2 × HD versus 5.1 ± 2.8 mL/min for 3 × HD: P = 0.507). In a mixed effects model correcting for between-group differences in comorbidities and demographics, 3×HD was associated with increased rate of loss of KRU and separation of KRU. In separate mixed effects models, group (2×HD versus 3×HD) was not associated with differences in serum potassium or phosphate, and the groups did not differ with respect to total standard Kt/V. Survival, adjusted for age, gender, weight, baseline KRU and comorbidity (prevalence of diabetes, cardiac disease, PVD and malignancy) was greater in the 2×HD group (hazard ratio 0.755: P = 0.044). In sub-analyses, the survival benefit was confined to women, and those of less than median bodyweight. Conclusion: 2×HD initiation as part of an incremental programme with regular monthly monitoring of KRU was safe and associated with a reduced rate of loss of RKF early after dialysis initiation and improved survival. Randomized controlled trials of this approach are indicated.