Background and objectives In dialysis patients, the associations between apoprotein profile and all-cause or cardiovascular disease (CVD)-related mortality are not well known. We, therefore, investigated whether apoprotein levels are associated with these events.Design, setting, participants, & measurements We undertook a prospective observational cohort study of prevalent hemodialysis patients aged $18 years (n=1081), who were followed for 4 years (2011)(2012)(2013)(2014). Outcomes were all-cause and CVD-related mortality. Predictors used were baseline apoprotein levels, particularly the apoprotein B (apo B)/ apoprotein A-1 (apo A-1) ratio. A Cox regression analysis was used to calculate the hazard ratios (HRs) for mortality. Apo A-1, apo B, and apo B/ apo A-1 ratio were analyzed with adjustments in three models: model 1, basic adjustment for age and sex; model 2, basic adjustments plus dialysis conditions (dialysis vintage, mean predialysis systolic blood pressure, dry weight, and mean intradialytic weight gain); and model 3, model 2 plus metabolic and inflammatory conditions (basal kidney disease, serum albumin, C-reactive protein level, and statin use).Results Of the 1081 patients included in the study, 203 deaths were recorded, 92 of which were related to CVD. The apo B/ apo A-1 ratio was significantly associated with all-cause and CVD-related mortality when analyzed by 1-SD increments or quartile IV versus I in all models. In model 3, HRs and 95% confidence intervals (95% CIs) for 1-SD increments of apo B/ apo A-1 ratio for all-cause mortality or CVD-related mortality were: HR, 1.16 (95% CI, 1.00 to 1.35), or HR, 1.38 (95% CI, 1.11 to 1.71), respectively, and for quartile IV versus I: HR, 1.65 (95% CI, 1.05 to 2.57), or HR, 2.56 (95% CI, 1.21 to 5.40), respectively. Apo A-1 was significantly associated with both mortalities in models 1 and 2. However, apo B was only significantly associated with CVD-related mortality in model 3.Conclusions Apoprotein measurement, especially the apo B/ apo A-1 ratio, was significantly associated with allcause and CVD-related mortality in prevalent dialysis patients.

No abstract

N-terminal-pro-B-type-natriuretic peptide (NT-proBNP) was a predictive marker of cardiovascular disease (CVD)-related death in chronic dialysis patients. NT-proBNP was also correlated with markers of inflammation, malnutrition and protein-energy wasting. We hypothesized whether NT-proBNP was also associated with non-CVD death in chronic dialysis patients. A prospective observational study for incidence of death in chronic dialysis patients was conducted. Prevalent chronic dialysis patients (n = 1310) were enrolled and followed for 24 months. One hundred forty-four deaths were recorded. Area under the curve using ROC analysis for NT-proBNP showed: all causes of death (0.761), CVD-related (0.750), infection and malignancy-related (0.702) and others and unknown (0.745). After adjusting for age, sex, hemodialysis vintage, cardiothoracic ratio, mean pre-dialysis systolic blood pressure, dry weight and basal kidney disease, the hazard ratios (95% confidence intervals) per 1-log NT-proBNP calculated using multivariate Cox analysis were: all causes of death, 3.83 (2.51–5.85); CVD-related, 4.30 (2.12–8.75); infection and malignancy-related, 2.41 (1.17-4.93); and others and unknown origin, 5.63 (2.57–12.37). NT-proBNP was significantly associated not only with CVD-relate but also with non-CVD-related deaths in this population of prevalent chronic dialysis patients.

BackgroundAlthough hemoglobin (Hb) levels are affected by a change in the body fluid status, the relationship between Hb levels and mortality while taking interdialytic weight gain (IDWG) at blood sampling into account has not yet been examined in hemodialysis patients.Study designCohort study.Setting, ParticipantsData from the Miyazaki Dialysis cohort study, including 1375 prevalent hemodialysis patients (median age (interquartile range), 69 (60–77) years, 42.3% female).PredictorPatients were divided into 5 categories according to baseline Hb levels and two groups based on the median value of IDWG rates at blood sampling at pre-HD on the first dialysis session of the week.OutcomesAll-cause and cardiovascular mortalities during a 3-year follow-up.MeasurementsHazard ratios were estimated using a Cox model for the relationship between Hb categories and mortality, and adjusted for potential confounders such as age, sex, dialysis duration, erythropoiesis-stimulating agent dosage, Kt/V, comorbid conditions, anti-hypertensive drug use, serum albumin, serum C-reactive protein, serum ferritin, and serum intact parathyroid hormone. Patients with Hb levels of 9–9.9 g/dL were set as our reference category.ResultsA total of 246 patients (18%) died of all-cause mortality, including 112 cardiovascular deaths. Lower Hb levels (<9.0g/dL) were associated with all-cause mortality (adjusted HRs 2.043 [95% CI, 1.347–3.009]), while Hb levels were not associated with cardiovascular mortality. When patients were divided into two groups using the median value of IDWG rates (high IDWG, ≥5.4% and low IDWG, <5.4%), the correlation between lower Hb levels and all-cause mortality disappeared in high IDWG patients, but was maintained in low IDWG patients (adjusted HRs 3.058 [95% CI,1.575–5.934]). On the other hand, higher Hb levels (≥12g/dL) were associated with cardiovascular mortality in high IDWG patients (adjusted HRs 2.724 [95% CI, 1.010–7.349]), but not in low IDWG patients.ConclusionIn hemodialysis patients, target Hb levels may need to be selected in consideration of IDWG at blood sampling.

No abstract