Ortality and CV mortality respectively. Table 3 shows the relative risks (hazard ratios) of death due to all causes and specific causes on univariate cox regression analysis. Patients on CAPD had a 62 increased risk of death from all causes as well as a greater than 2-fold increase in the risk of death from infectious causes. Significantly as well, there was a 10 reduction in the risk of death (all-cause mortality) for every unit increase in haemoglobin levels. Although it was not of statistical significance, patients with DM had almost twice the risk of death from CV causes in comparison to GSK089 dose non-diabetics. Weight at initiation of dialysis was however significantly associated with the risk of CV death (HR: 0.97, CI: 0.95?.99, p = 0.04). In assessing potential baseline predictors of all-cause mortality, CAPD remained an independent predictor of all-cause mortality (HR: 2.00, CI: 1.29?.10) after adjusting for weight atTable 3. Univariate Hazard ratios for all-cause and cause-specific mortality. Characteristic Modality (PD) Age (years) Gender (male) Distance to order MLN9708 centre (50 Km) Housing (Informal) Diabetes mellitus (present) Hypertension (present) Weight (kg) sAlbumin (g/L) sCholesterol (mmol/L) Corrected Calcium (mmol/L) CXP (mmol2/L2) sHemoglobin (g/dl) sFerritin EPO (per 1000units) * p < 0.05 sAlbumin--serum albumin; sCholesterol--serum cholesterol--CXP--Calcium-phosphate product.; sHaemoglobin--serum haemoglobin; sFerritin--serum ferritin; EPO--Erythropoietin doi:10.1371/journal.pone.0156642.t003 All-cause Mortality 1.62 (1.07?.46)* 1.00 (0.98?.-02) 0.89 (0.59?.34) 1.00 (0.99?.00) 1.28 (0.82?.03) 1.43 (0.79?.56) 0.88 (0.54?.43) 0.99 (0.98?.00) 0.98 (0.99?.01) 1.03 (0.86?.23) 1.25 (0.61?.57) 0.86 (0.75?.99)* 0.90 (0.82?.99)* 1.00 (0.99?.0007) 0.99 (0.95?.04) CV mortality 1.34 (0.62?.88) 1.02 (0.99?.06) 0.83 (0.39?.78) 1.00 (0.99?.01) 1.30(0.54?.25) 1.86(0.73?.80) 0.84 (0.34?.09) 0.97 (0.95?.99)* 0.99 (0.94?.06) 0.93 (0.66?.30) 2.46 (0.49?2.23) 0.71 (0.55?.90)* 0.90 (0.75?.07) 1.00 (0.99?.001) 0.99 (0.92?.07) Infection-related mortality 2.27 (1.13?.60)* 1.00 (0.97?.03) 1.24 (0.61?.51) 1.09 (0.88?.35) 1.32 (0.61?.85) 1.62 (0.62?.23) 1.24 (0.52?.67) 1.01 (0.99?.02) 0.98 (0.93?.04) 1.08 (0.68?.72) 0.95 (0.25?.63) 0.84 (0.68?.03) 0.93(0.77?.14) 1.00 (0.99?.00) 0.96 (0.89?.03)PLOS ONE | DOI:10.1371/journal.pone.0156642 June 14,7 /Baseline Predictors of Mortality in Chronic Dialysis Patients in Limpopo, South AfricaTable 4. Multivariate Cox regression model of the baseline predictors of all-cause mortality. Characteristic Modality (PD) Diabetes mellitus (present) Hypertension (present) sHemoglobin (g/dl) sAlbumin (g/L) Weight (kg) Hazard ratio 2.00 1.67 0.86 0.87 0.98 0.98 Confidence intervals 1.29?.10 0.92?.02 0.52?.41 0.79?.97 0.97?.00 0.97?.00 p-value 0.002 0.09 0.55 0.01 0.25 0.sAlbumin--serum albumin; sHaemoglobin--serum haemoglobin; sFerritin doi:10.1371/journal.pone.0156642.tdialysis commencement, comorbidity (diabetes mellitus), anaemia (using haemoglobin as a measure), and baseline serum albumin(Table 4). Likewise, baseline haemoglobin remained a predictor of all-cause mortality with a 13 reduction in risk of death from all-causes for every unit increase in haemoglobin levels (HR: 0.87 CI: 0.79?.97, p = 0.01). Mortality risk according dialysis modality was significantly modified by diabetes mellitus status on both univariate and multivariable analyses (Table 5). Mortality risk was approximately 5 times higher among diabetics who had CA.Ortality and CV mortality respectively. Table 3 shows the relative risks (hazard ratios) of death due to all causes and specific causes on univariate cox regression analysis. Patients on CAPD had a 62 increased risk of death from all causes as well as a greater than 2-fold increase in the risk of death from infectious causes. Significantly as well, there was a 10 reduction in the risk of death (all-cause mortality) for every unit increase in haemoglobin levels. Although it was not of statistical significance, patients with DM had almost twice the risk of death from CV causes in comparison to non-diabetics. Weight at initiation of dialysis was however significantly associated with the risk of CV death (HR: 0.97, CI: 0.95?.99, p = 0.04). In assessing potential baseline predictors of all-cause mortality, CAPD remained an independent predictor of all-cause mortality (HR: 2.00, CI: 1.29?.10) after adjusting for weight atTable 3. Univariate Hazard ratios for all-cause and cause-specific mortality. Characteristic Modality (PD) Age (years) Gender (male) Distance to centre (50 Km) Housing (Informal) Diabetes mellitus (present) Hypertension (present) Weight (kg) sAlbumin (g/L) sCholesterol (mmol/L) Corrected Calcium (mmol/L) CXP (mmol2/L2) sHemoglobin (g/dl) sFerritin EPO (per 1000units) * p < 0.05 sAlbumin--serum albumin; sCholesterol--serum cholesterol--CXP--Calcium-phosphate product.; sHaemoglobin--serum haemoglobin; sFerritin--serum ferritin; EPO--Erythropoietin doi:10.1371/journal.pone.0156642.t003 All-cause Mortality 1.62 (1.07?.46)* 1.00 (0.98?.-02) 0.89 (0.59?.34) 1.00 (0.99?.00) 1.28 (0.82?.03) 1.43 (0.79?.56) 0.88 (0.54?.43) 0.99 (0.98?.00) 0.98 (0.99?.01) 1.03 (0.86?.23) 1.25 (0.61?.57) 0.86 (0.75?.99)* 0.90 (0.82?.99)* 1.00 (0.99?.0007) 0.99 (0.95?.04) CV mortality 1.34 (0.62?.88) 1.02 (0.99?.06) 0.83 (0.39?.78) 1.00 (0.99?.01) 1.30(0.54?.25) 1.86(0.73?.80) 0.84 (0.34?.09) 0.97 (0.95?.99)* 0.99 (0.94?.06) 0.93 (0.66?.30) 2.46 (0.49?2.23) 0.71 (0.55?.90)* 0.90 (0.75?.07) 1.00 (0.99?.001) 0.99 (0.92?.07) Infection-related mortality 2.27 (1.13?.60)* 1.00 (0.97?.03) 1.24 (0.61?.51) 1.09 (0.88?.35) 1.32 (0.61?.85) 1.62 (0.62?.23) 1.24 (0.52?.67) 1.01 (0.99?.02) 0.98 (0.93?.04) 1.08 (0.68?.72) 0.95 (0.25?.63) 0.84 (0.68?.03) 0.93(0.77?.14) 1.00 (0.99?.00) 0.96 (0.89?.03)PLOS ONE | DOI:10.1371/journal.pone.0156642 June 14,7 /Baseline Predictors of Mortality in Chronic Dialysis Patients in Limpopo, South AfricaTable 4. Multivariate Cox regression model of the baseline predictors of all-cause mortality. Characteristic Modality (PD) Diabetes mellitus (present) Hypertension (present) sHemoglobin (g/dl) sAlbumin (g/L) Weight (kg) Hazard ratio 2.00 1.67 0.86 0.87 0.98 0.98 Confidence intervals 1.29?.10 0.92?.02 0.52?.41 0.79?.97 0.97?.00 0.97?.00 p-value 0.002 0.09 0.55 0.01 0.25 0.sAlbumin--serum albumin; sHaemoglobin--serum haemoglobin; sFerritin doi:10.1371/journal.pone.0156642.tdialysis commencement, comorbidity (diabetes mellitus), anaemia (using haemoglobin as a measure), and baseline serum albumin(Table 4). Likewise, baseline haemoglobin remained a predictor of all-cause mortality with a 13 reduction in risk of death from all-causes for every unit increase in haemoglobin levels (HR: 0.87 CI: 0.79?.97, p = 0.01). Mortality risk according dialysis modality was significantly modified by diabetes mellitus status on both univariate and multivariable analyses (Table 5). Mortality risk was approximately 5 times higher among diabetics who had CA.
