Design: randomised, two-period, two-sequence crossover trial in two outpatient units
Population: 36 patients randomised on maintenance haemodialysis, 32 completing both periods
Primary outcome: dialysis adequacy by urea reduction ratio and single-pool Kt/V, and fatigue by the Piper Fatigue Scale
Effect: no significant difference in adequacy or fatigue; greater intradialytic systolic and diastolic blood pressure reductions with feeding, most marked in the second hour

Thirty-six patients on maintenance haemodialysis across two outpatient units in Turkiye were randomised in a two-period, two-sequence crossover design, 32 completing both periods. In one period they received standardised intradialytic nutrition; in the other, no food during the session. Adequacy was measured by urea reduction ratio and single-pool Kt/V, fatigue by the Piper Fatigue Scale, with intradialytic blood pressure recorded throughout.

Neither dialysis adequacy nor fatigue differed between periods. What did differ was haemodynamics: feeding was associated with greater falls in both systolic and diastolic pressure, most marked during the second hour. No clinically relevant adverse events occurred.

This addresses a real disagreement between units. Feeding during dialysis is restricted in many centres on the grounds that splanchnic blood diversion reduces clearance and provokes hypotension; others encourage it because these patients are chronically undernourished and the session is three hours of otherwise unused time. This trial supports the second position on adequacy and the first on haemodynamics. The sensible synthesis is to permit intradialytic feeding, and to time it away from the point in the session where ultrafiltration is steepest, with the blood pressure watched rather than assumed.

Do not restrict intradialytic feeding on the grounds that it reduces clearance - adequacy was unaffected
Watch blood pressure through the second hour in patients who eat during dialysis
Time feeding earlier in the session, away from the steepest ultrafiltration
Individualise for patients already prone to intradialytic hypotension
Weigh the nutritional gain explicitly - protein-energy wasting is the commoner problem in this population

Why it matters

It settles the clearance half of a long-running disagreement about whether patients should eat during dialysis.

The statistics, in plain English

This is a small crossover trial, 32 completers, and its null findings on adequacy and fatigue should be read as no signal rather than as proof of no difference; with this sample a modest reduction in Kt/V could have been missed. The crossover design is a genuine strength, because each patient serves as their own control and removes between-patient variation. The blood pressure finding is reported as an association without effect sizes in the abstract, so the magnitude of the fall is not quantified here and should not be guessed at.