Fujisaki K, Tanaka S, Taniguchi M, Matsukuma Y, Masutani K, Hirakata H, Kitazono T, Tsuruya K: Study on Dialysis Session Length and Mortality in Maintenance Hemodialysis Patients: The Q-Cohort Study. Nephron 10.1159/000489680
The dialysis treatment session length (DSL) in conventional thrice weekly hemodialysis therapy has been long suspected to contribute importantly to the “adequacy” of therapy for end stage renal disease (ESRD) independent of dose of therapy (spKt/V urea). However, the independent impact of DSL on patient centered outcomes (such as survival and hospitalization rates) has been difficult to quantify by randomized trials. Nevertheless, short DSL (less than 3 – 4 hours) has been suspected to represent “inadequate” treatment, especially in those patients with high inter-dialytic weight gain mandating high intra-dialytic ultrafiltration rates (often exceeding 13 ml/kg/hour), frequently observed in oliguric subjects with little residual renal function.
Upon this background of uncertainty, Fujisaki and co-workers report on a prospective, propensity-matched cohort study (“Q Cohort Study”) conducted in Japan between 2006 and 2010 examining the risk factor of DSL and its association with all-cause mortality (ACM) in conventional thrice weekly hemodialysis. They studied 3486 prevalent hemodialysis patients (average age 64 years, dialysis vintage 7.7 years) followed for 4 years. In a propensity match sub-cohort (n = 1934), a DSL of 4.1 hours was compared to a DSL of 5.1 hours. Subjects with DSL of 5 hours or more had a slightly higher spKt/V urea of 1.6 ± 0.3 versus 1.5 ± 0.3 for those with a DSL of < 5 hours. Unfortunately, no data is available on residual renal function so the study may be biased by treatment indication as this was not a randomized trial. Nevertheless, in a propensity-matched, fully adjusted analysis the hazard rate (HR) for ACM was 22 % lower in the long-DSL cohort (p = 0.12). The impact of long-DSL was more pronounced in the very elderly (> 80 years) and not significant in subjects younger than 80 years. It needs to be emphasized that this is a study of DSL in prevalent but not incident hemodialysis patients. In addition, only ACM not cardiovascular mortality was examined. Earlier studies have largely supported the concept that short DSL (< 4 hours) with conventional thrice weekly treatment sessions can be associated with an increased risk of mortality when applied in incident patients, especially with low values for residual renal function or oliguria. But randomized clinical trials, usually with DSA of < 4.5 hours, have generally failed to demonstrate beneficial effects of longer DSL, but some of these studies may have been underpowered to show such effects or have confounding variables affecting interpretation. It does seem fairly clear that short-DSL obligate higher intra-dialytic ultrafiltration rates that can have a deleterious effect on long term survival.
While supplying mostly moderately weak evidence, this study does support the notion that longer DSL is better (at least for the elderly prevalent dialysis patient. But more well-powered randomized trials in subjects with minimal residual renal function are still needed to determine the optimal DSL in thrice weekly conventional hemodialysis, both in incident and prevalent patients of varying age and dialysis vintage. In the meantime, in my opinion, any DSL of < 4 hours with a thrice weekly schedule of treatments should not be regarded as “adequate” in a patient with ESRD and minimal residual renal function, independent of spKt/V.