To prevent protein energy malnutrition (PEM) and accumulation of waste products, dialysis patients require diet adjustments. Dietary intake assessed by self-reported intakes often provides biased information and standard 24-h urinary excretion is inapplicable in dialysis patients. We aimed to assess dietary intake via a complementary, less biased biomarker method, and to compare this to dietary diaries. Additionally, we investigated the prospective association of creatinine excretion rate (CER) reflecting muscle mass with mortality. Complete intradialytic dialysate and interdialytic urinary collections were used to calculate 24-h excretion of protein, sodium, potassium, phosphate and creatinine in 42 chronic dialysis patients and compared with protein, sodium, potassium, and phosphate intake assessed by 5-day dietary diaries. Cox regression analyses were employed to investigate associations of CER with mortality. Mean age was 64 +/- 13 years and 52% were male. Complementary biomarker assessed (CBA) and dietary assessed (DA) protein intake were significantly correlated (r = 0.610; p <0.001), but there was a constant bias, as dietary diaries overestimated protein intake in most patients. Correlations were found between CBA and DA sodium intake (r = 0.297; p = 0.056), potassium intake (r = 0.312; p = 0.047) and phosphate uptake/intake (r = 0.409; p = 0.008). However, Bland-Altman analysis showed significant proportional bias. During a median follow-up of 26.6 (25.3-31.5) months, nine dialysis patients (23%) died. CER was independently and inversely associated with survival (HR: 0.59 (0.42-0.84); p = 0.003). Excretion measurements may be a more reliable assessment of dietary intake in dialysis patients, as this method is relatively free from biases known to exist for self-reported intakes. CER seems to be a promising tool for monitoring PEM.