Haemodiafiltration likely lowers death compared with haemodialysis, says Cochrane update

Design

Cochrane systematic review and meta-analysis of randomised trials

Population

62 trials, 6,897 adults with kidney failure on maintenance dialysis

Primary outcome

All-cause and cardiovascular death

Effect

All-cause death RR 0.88 (0.80 to 0.96); cardiovascular death RR 0.78 (0.66 to 0.92)

This is the third version of the Cochrane review, searched to April 2025. It includes 62 randomised trials and 6,897 adults with kidney failure, most comparing convective therapies — haemodiafiltration, haemofiltration or acetate-free biofiltration — with high- or low-flux haemodialysis.

Convective dialysis likely reduced all-cause death (RR 0.88, 95% CI 0.80 to 0.96; 15 trials, 5,463 participants) and cardiovascular death (RR 0.78, 0.66 to 0.92; 9 trials), with no heterogeneity and moderate certainty, downgraded for indirectness. Non-fatal cardiovascular events did not differ (RR 1.05). Evidence on fatigue, vascular access, admissions and quality of life was too sparse and imprecise to conclude.

A 12% relative reduction in death is meaningful in a population with high mortality. The benefit rests on survival rather than on how patients feel, and adoption depends on machines, ultrapure water and cost — real constraints in many Indian units. Where haemodiafiltration is available, this supports it as the default for suitable patients.

• Consider haemodiafiltration as the default modality where equipment and ultrapure water are available
• Discuss the likely survival benefit with patients choosing between modalities
• Do not promise better fatigue or quality of life; that evidence is inconclusive
• Plan water treatment and monitoring standards before introducing haemodiafiltration in a unit
• Keep optimising dialysis adequacy, fluid balance and access care whatever the modality

Why it matters

Moderate-certainty evidence of lower mortality is rare in dialysis, and it points to a modality choice units can act on.

Don't overread it

The benefit is in survival; effects on symptoms, admissions and quality of life remain unproven.

The statistics, in plain English

A risk ratio of 0.88 means about 12% fewer deaths. With no heterogeneity (I² = 0%) across 15 trials, the result is consistent, and certainty was marked down only because the trials did not perfectly match everyday practice.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.