Nuwellis, Inc. announced results from two new clinical data analyses from the AVOID-HF trial, which demonstrate the benefits of its Aquadex System in reducing heart failure readmissions at 30 days. Results from the analyses were presented in a late-breaking session at the Technology and Heart Failure Therapeutics (THT) conference in Boston. The data was presented by Dr. Sean P. Pinney, M.D., Professor of Medicine and director of the Heart Failure and Transplantation Program at the Mount Sinai Health System, and evaluated the clinical benefit of adjustable ultrafiltration (AUF) therapy with Nuwellis?

Aquadex System compared to adjustable loop diuretics (ALD) by re-analyzing data from the AVOID-HF (Aquapheresis vs. Intravenous Diuretics and Hospitalizations for Heart Failure) clinical trial1 using the Finkelstein-Schoenfeld method of Win-Ratios. In the trial, 221 study participants were randomized to AUF (n=110) or ALD (n=111), and 213 (AUF=105, AUD= 108) patients who completed index treatment and discharge were included in the analysis.

Data were independently adjudicated by a blinded clinical events committee, which evaluated AUF compared to ALD within the three-factor composite endpoint of cardiovascular mortality, heart failure events and quality of life. Key findings demonstrating the benefits of AUF include: Fewer heart failure events and heart failure hospitalizations: AUF patients had significantly fewer heart failure events within 30 days compared to ALD (90% vs 77.3% p=0.0138) and fewer heart failure hospitalizations for the AUF patients compared to the ALD patients (90.0% vs. 79.2% p=0.0321) within 30 days.

Results of the Hierarchical Win-Ratio: In the primary composite outcome, 72.6% resulted in either a ?win? or ?loss? and the remaining 27.4% resulted in a ?tie?.

AUF won in 71.0% of the heart failure event related paired comparisons (versus 29.0% for ALD) and in 53.4% of the quality-of-life comparisons (versus 46.6% for ALD) resulting in a WR=1.43 (p=0.056) favoring ultrafiltration. Other statistically significant results presented from the original AVOID-HF trial1 included: Fewer patients re-hospitalized for heart failure (p=0.034) Fewer days in the hospital due to heart failure readmissions (p=0.029) Lower rehospitalization rates due to a cardiovascular event (p=0.037) Fewer rehospitalization days due to a cardiovascular event (p=0.018) Fewer patients re-hospitalized for a cardiovascular event (p=0.042) This study, presented by Deya Alkhatib, M.D., Section of Cardiovascular Medicine, Yale School of Medicine, aimed to develop a model for pretreatment and identification of risk for 90-day heart failure events among heart failure patients who have undergone AUF therapy. Using artificial intelligence (AI) and machine learning (ML), a predictive model was developed based on data from the AVOID-HF trial.

The model was designed to be used before initiating AUF to anticipate which patients will respond well to the therapy and which will be at high risk for future heart failure events. Key findings from the analysis include: Top predictors for 90-day heart failure events: Using ML, the study identified the top 10 predictors for 90-day heart failure events. Notably, ?intimate relationships with loved ones?

was a strong predictor of response to AUF therapy. Other predictors included valvular heart disease, history of arrhythmia, poor adherence to medical therapy, history of diabetes mellitus, suboptimal diuretic therapy response, chronic obstructive lung disease, ALD use during acute decompensated heart failure hospitalization, history of cerebrovascular disease, and intravenous bumetanide use. Successful prediction of outcomes: The ML model used in the study was more successful in predicting the outcome for heart failure patients treated with AUF.

The predictive model anticipated 90-day heart failure events with better statistical accuracy than existing classic models. Strong results for super-responders: 90% of patients categorized as super-responders to AUF therapy within this model did not experience any 90-day heart failure events. Accurate predictions for high-risk patients: The model assigned 41% of patients in the study to the high-risk category.

Among these patients, 57% experienced a 90-day heart failure event.