RESUMO
The CONVINCE trial demonstrates that high-dose hemodiafiltration offers a survival advantage for patients in the high-flux hemodiafiltration group compared to hemodialysis. We compared the outcomes of hemodialysis and hemodiafiltration using real-world data. We conducted an analysis on a cohort of patients who underwent hemodiafiltration therapy (HDF) at a single center, NefroStar Clinics. The results obtained were then compared with data from patients receiving hemodialysis (HD) therapy within the Brazilian Public Health System (SUS). The primary outcome was mortality from any cause. Results: A total of 85 patients undergoing hemodiafiltration were compared with 149,372 patients receiving hemodialysis through the Brazilian Public Health System (SUS). Using a 2:1 propensity score, we compared the 170 best-match HD patients with 85 HDF patients. In the Cox analysis, HDF therapy showed a reduced risk of mortality with an HR of 0.29 [0.11-0.77]. The propensity score analysis showed a HR of 0.32 [95% CI: 0.11-0.91]. This analysis was adjusted for age, type of access, KT/v, hemoglobin, and phosphorus. The Kaplan-Meier analysis showed respective survival rates for HDF and HD at the end of one year, 92.1% and 79.9%, p < 0.001. These results suggest high-flux hemodiafiltration has survival advantages over hemodialysis in a real-world scenario.
RESUMO
Elderly is the main age group affected by acute kidney injury (AKI). There are no studies that investigated the predictive properties of urinary (u) NGAL as an AKI marker in septic elderly population. This study aimed to evaluate the efficacy of uNGAL as predictor of AKI diagnosis and prognosis in elderly septic patients admitted to ICUs. We prospectively studied elderly patients with sepsis admitted to ICUs from October 2014 to November 2015. Assessment of renal function was performed daily by serum creatinine and urine output. The level of uNGAL was performed within the first 48 hours of the diagnosis of sepsis (NGAL1) and between 48 and 96 hours (NGAL2). The results were presented using descriptive statistics and area under the receiver operating characteristic curve (AUC-ROC) and p value was 5%. Seventy-five patients were included, 47 (62.7%) developed AKI. At logistic regression, chronic kidney disease and low mean blood pressure at admission were identified as factors associated with AKI (OR=0.05, CI=0.01-0.60, p=0.045 and OR=0.81, CI=0,13-0.47; p=0.047). The uNGAL was excellent predictor of AKI diagnosis (AUC-ROC >0.95, and sensitivity and specificity>0.89), anticipating the AKI diagnosis in 2.1±0.3 days. Factors associated with mortality in the logistic regression were presence of AKI (OR=2.14, CI=1.42-3.98, p=0.04), chronic obstructive pulmonary disease (OR = 9.37, CI =1.79-49.1, p=0.008) and vasoactive drugs (OR=2.06, CI=0.98-1.02, p=0.04). The accuracy of NGALu 1 and 2 as predictors of death was intermediate, with AUC-ROC of 0.61 and 0.62; sensitivity between 0.65 and 0.77 and specificity lower than 0.6. The uNGAL was excellent predictor of AKI in septic elderly patients in ICUs and can anticipate the diagnosis of AKI in 2.1 days.
