MO881
LUNG ULTRASONOGRAPHY AND BIOIMPEDANCE IN ASSESSMENT OF VOLUME STATUS OF HEMODIALYSIS PATIENTS: A COMPARISON BETWEEN BOTH METHODS

Assessment of volume status is an important prognostic factor in hemodialysis (HD) patients. Several methods have been suggested to estimate it: bioimpedance analysis, brain natriuretic peptide levels (BNP) and lung and inferior vena cava (IVC) ultrasonography (US), which are emerging as a valuable technique in this field. Our aim was to evaluate effectiveness of lung US in assessment of volume status in chronic HD patients and compare it with the gold standard bioimpedance technique.

Cross-sectional study of 58 prevalent HD patients. Several analytical data were analyzed, including BNP, albumin and sodium levels. Lung and IVC US were performed to assess the presence and distribution of B-lines and the diameter and respiratory collapsibility of IVC, respectively. Fresenius® body composition monitor (BCM) was the bioimpedance technique used. It defines hyperhydration as relative fluid overload (RFO) >15%. Both US and BCM were performed at the same day, immediately before the middle week HD session. Categorical variables are presented as frequencies and percentages, continuous variables as means and standard deviations, or medians and interquartile ranges (IQR) for variables with skewed distributions. Statistical analysis was performed using SPSS® version 25 for Windows.

The average age was 75.3±1.6 years and 56.9% were male. The average time in HD was 36.6±4.1 months. Half of the patients (n=29) presented B-lines in lung US. The diameter of IVC was significantly higher in this group of patients, both inhaling (1.3±0.4 vs 0.9±0.4; p=0.001) and exhaling (2.0±0.4 vs 1.6±0.5; p<0.001). Similarly, most patients with B-lines at lung US presented collapsibility of IVC < 50% (n=25; 89.3%).

Although not statistically significant, the RFO was higher in patients with presence of B-lines compared to those who did not present it (10.2±10.6% vs 5.4±6.9%; p=0.061). Considering the mean RFO value in patients with B-lines, we observed that its presence was significantly higher in patients with RFO > 10% compared to those with RFO < 10% (58.6% vs 31%; p=0.035). Besides, patients with RFO > 10% also presented higher BNP (9878±34646 pg/mL vs 3945±8634 pg/mL; p=0.009) and lower albumin levels (3.4±0.5 vs 3.7±0.4; p= 0.039) than other group.

Lung US seems to be a valuable tool to diagnose overhydration earlier than BCM, which currently defines hyperhydration as RFO >15%. In this study we verified that patients with lower levels of RFO (>10%) already presented signals of fluid overload, such as presence of B-lines at lung US, higher BNP and lower albumin levels. These findings support the benefit of lung US in guiding fluid removal and may change the practice in our HD unit to probe the ideal dry weight for chronic dialysis patients.