Early as compared to late initiation of twice-weekly hemodialysis and short-term survival among end-stage renal disease patients.

Introduction: The impact of timing of hemodialysis (HD) for end-stage renal disease (ESRD) patients treated with twice-weekly HD remains unclear. We aimed to determine the effects of late initiation of HD on short-term mortality and hospitalization.

Methods: A multicenter cohort study was conducted in 11 HD centers in Northeastern Thailand (HEmodialysis Network of the NorthEastern Thailand study group). We recruited adult ESRD patients who were treated with twice-weekly HD for more than 3 months and had data on eGFR at HD initiation. Clinical and laboratory values at the time of recruitment were recorded. Late and early (eGFR at start <5 and >5 ml/min/1.73 m ) initiations were defined. Outcomes were disease-related death (excluding any accidental deaths) and first hospitalization. Data analysis was performed by multivariable cox-regression analysis.

Findings: A total of 407 patients who had data on eGFR at HD initiation (303 in late group and 104 in early group) were included for analysis. There were 56.8% male with a mean age of 55 years. During the 15.1 months of follow-up, there were 27 (6.6%) disease-related deaths. The 1-year survival rate was similar among late and early initiation groups. The incidence density of first hospitalization in the late group was significantly lower than those in the early group (HR adjusted, 0.63; 95% CI, 0.40-0.99, p = 0.047). Among 303 patients who were in the late start group, patients with diabetes had a higher mortality rate (HR, 3.49; 95% CI, 1.40-8.70, p = 0.007) when compared to non-diabetic patients.

Discussion: Early initiation of HD at eGFR >5 ml/min/1.73 m had no short-term survival benefit compared to the late group in ESRD patients treated with twice-weekly HD for at least 3 months in a resource-limited setting. A survival benefit from an early start of HD was found among diabetic patients.