Abstract Background and Aims Correct estimation of the fluid status as represented by the “Dry body weight” (DBW) of hemodialysis (HD) patients is crucial for their cardiovascular health and general well-being. However, DBW assessment through medical history and physical examination by the attending nephrologist may not prove sufficiently accurate. Bioimpedance measurements may offer a more accurate assessment of DBW. Therefore we compared clinical DBW estimation with DBW estimation by bioimpedance. We searched for variables which may correlate with differences between the two methods. Method Observational, single center cross sectional study. Fifty HD patients (32 males) consented to participate. Laboratory, clinical and epidemiological data were drawn from the medical records. Bioimpedance measurements were performed by the Body Composition Monitor (Fresenius Medical Care) before a mid- week hemodialysis session and were compared with parallel DBW estimations based on medical history and physical examination. A variable “difference DBW” was derived from the difference between the two parallel estimations of DBW (“difference DBW” = clinical DBW estimation – bioimpedance DBW measurement). Univariate and bivariate correlation analyses were used. Results The mean age, years in end stage renal disease (ESRD), Kt/V, serum albumin, body mass index (BMI), lean tissue index (LTI), and adipose tissue mass (ATM) of the participants were 66.8±13.878y, 8±8.132y, 1.48±0.269, 3.8±0.313g/dL, 27.97±4.595, 13.26±3.982, 37.32±13.159Kg respectively. DBW estimated by the nephrologists was significantly lower (72.94 ± 14.637 kg) than DBW estimated by bioimpedance (74.82 ± 15.594 kg), a decrease of 1.882 (95% CI, -2.425 to -1.339kg), t(49) = -6.963, p < 0.001. Significant negative correlation was found between “difference DBW” and BMI r(48) =-0.69, p<0.001. Moderate negative correlation was found between “difference DBW” and LTI r(48) = -0.42, p=0.002, and between “difference DBW” and ATM, r(48) = -0.41, p=0.003. Moderate positive correlation was found between “difference DBW” and years in ESRD r(48) = 0.43, p=0.002. There were no significant correlations between sex, age and “difference DBW”. Conclusion In our cohort there were significant differences between clinical DBW estimations and corresponding Bioimpedance DBW measurements. This difference was more pronounced in patients with low BMI and ATM, as well as in patients with more years on ESRD. Therefore we consider that these findings should be taken into consideration when determining DBW. Figure SEQ Εικόνα \* ARABIC 1 Scatterplots (A) Significant negative correlation between “difference DBW” and BMI r(48) =-0.69, p<0.001. (B) Moderate negative correlation between “difference DBW” and ATM, r(48) = -0.41, p=0.003. (C) Moderate positive correlation between “difference DBW” and years in ESRD r(48) = 0.43, p=0.002. DBW: dry body weight, BMI: body mass index, ATM: adipose tissue mass, ESRD: end stage renal disease