OBJECTIVE Continuous veno-venous hemofiltration (CVVH) is a mode of renal replacement therapy in critically ill patients that has gained popularity all over the world. The authors reviewed one-year experience with CVVH in intensive care units (ICUs) of Bhumibol Adulyadej Hospital. The objectives of this study were to describe the characteristics of the patients and demonstrate the association between various factors and outcome. METHODS The medical records of 45 patients who underwent CVVH treatment were analyzed. All patients had been admitted into the ICUs of Bhumibol Adulyadej Hospital between 1 January 2005 and 31 December 2005. RESULTS Average age of patients was 67.7 +/- 13.3 years (range from 27.0 years to 88.4 years). The male: female ratio was 1.4:1. Twenty-four patients were admitted to the medical ICU, 17 to the coronary care unit (CCU) and 4 to the surgical ICU. All of them needed mechanical ventilator support and 91.1% required vasopressor. Sixty percent of the patients had sepsis. Most of them had comorbidity including, cardiovascular (66.7%), hepatobilliary (35.6%) and neurological comorbidity (13.3%). Half of them had been diagnosed with chronic kidney disease (CKD) with pre-dialysis CKD in 40% and end stage renal disease (ESRD) in 11.1%. The mean number of organ failure was 3.18 +/- 0.1 and 95.6% had more than 2 organ failures. The range of APACHE II score was 15-50 (mean 30.8 +/- 9.5) with a predicted death rate of 21.0-97.8% (mean 66.4 +/- 23.4). The indications for renal replacement were 80% for level of nitrogenous waste product, 75.6% for volume overload, 42.2% for severe metabolic acidosis, 35.6% for hyperkalemia and 2.2% for toxic substance removal. Eighty percent of the patients had 2 or more indications. Mean blood urea nitrogen (BUN) and creatinine (Cr) level before starting CVVH was 78.8 +/- 36.5 mg/dl (10.0 to 187.0) and 5.3 +/- 3.3 mg/dl (2.0 to 20.2) respectively. Duration of CVVH was 1.5 to 251.0 hours (mean 57.8 +/- 58.9) and the average CVVH dose was 36.6 +/- 7.5 ml/kg/hr (24.6 to 55.6). The overall mortality was 80%. The two most frequent causes of death were sepsis (44.4%) and cardiovascular disease (15.6%). The significant difference between the survival and nonsurvival groups were surgical ICU admission (p = 0.021), sepsis (p = 0.019), APACHE II score (p = 0.011), volume indication for CVVH (p = 0.028), number of dialysis indication (p = 0.019), duration of hospitalization (p = 0.004), systolic blood pressure (p = 0.012) and serum albumin level (p = 0.009). By logistic regression analysis, there was only statistical significance for serum albumin level less than 3 g/dl. CONCLUSIONS One-year experience of CVVH in Bhumibol Adulyadej Hospital showed high mortality rate but it is comparable to previous publications. The only factor that was associated with death by multivariate analysis was lower serum albumin level at the time of initiating CVVH. Prospective studies are required to explore this issue in the future.