Registry on extracorporeal multiple organ support with the advanced organ support (ADVOS) system: 2-year interim analysis

The objective of this registry is to collect data on real-life treatment conditions for patients for whom multiple organ dialysis with Advanced Organ Support (ADVOS) albumin hemodialysis is indicated. This registry was performed under routine conditions and without any study-specific intervention, diagnostic procedures, or assessments. Data on clinical laboratory tests, health status, liver function, vital signs, and examinations were collected (DRKS-ID: DRKS00017068). Mortality rates 28 and 90 days after the first ADVOS treatment, adverse events and ADVOS treatment parameters, including treatment abortions, were documented. This analysis was performed 2 years after the first patient was included on January 18, 2017. As of February 20, 2019, 4 clinical sites in Germany participated and enrolled 118 patients with a median age of 60 (IQR: 45, 69) of whom 70 were male (59.3%). Patients had a median SOFA Score of 14 (IQR: 11, 16) and a predicted mortality of 80%. The median number of failing organs was 3 (IQR: 2, 4). Four hundred twenty nine ADVOS treatments sessions were performed with a median duration of 17 hours (IQR: 6, 23). A 5.8% of the ADVOS sessions (25 of 429) were aborted due to device related errors, while 14.5% (62 of 429) were stopped for other reasons. Seventy nine adverse events were documented, 13 of them device related (all clotting, and all recovered without sequels). A significant reduction in serum creatinine (1.5 vs 1.2 mg/dl), blood urea nitrogen (24 vs 17 mg/dl) and bilirubin (6.9 vs 6.5 mg/dl) was observed following the first ADVOS treatment session. Blood pH, bicarbonate (HCO(3)(-)) and base excess returned to the physiological range, while partial pressure of carbon dioxide (pCO(2)) remained unchanged. At the time of the analysis, 28- and 90-day mortality were 60% and 65%, respectively, compared to an expected ICU-mortality rate of 80%. SOFA score was an independent predictor for outcome in a multivariable logistic regression analysis. The reported data show a high quality and completion of all participating centers. Data interpretation must be cautious due to the small number of patients, and the nature of the registry, without a control group. However, the data presented here show an improvement of expected mortality rates. Minor clotting events similar to other dialysis therapies occurred during the treatments.