Objective and Background: Catheter-based treatment of patients with ventricular arrhythmias (VA) reduces VA and mortality in selected patients. Data on optimal patient selection is scarce. The Risk in Ventricular Ablation (RIVA) Scorewas developed to assess the risk for major postinterventional complications in patients undergoing VA ablation. We sought to validate this score and to test for possible additional predictors.

Methods and Results: We analyzed 1964 catheter ablations for VT in patients with (n=1069; 54.4%) and without (n=893, 45.6%) structural heart disease and observed an overall major adverse event rate of 4.0% with an in-hospital mortality of 1.3% with significantly less complications occurring in patients without structural heart disease (1.1% vs. 6.5%; p < 0.01). The RIVA-Score, a recently published scoring system that uses underlying heart disease, kidney function, previous heart surgery, oral anticoagulation, and epicardial access demonstrated to be a valid predictive tool for major-in hospital complications (OR: 1.18; 95% CI: 1.12, 1.25; p ≤ 0.001).  Additionally, our multivariate logistic regression analysis demonstrated the presence of heart failure with a NYHA-Class ≥III (OR: 2.5; 95 % CI: 1.5, 4.2; p < 0.001) and age (OR: 1.04; 95 % CI: 1.02, 1.07; p < 0.001) to be predictive parameters. In addition to significance analysis, we evaluated the importance of features by considering permutation weight scores. With this machine learning technique, we distinguished between parameters that contain information for the calculation of the risk score and those that complicate the model without providing any meaningful contribution. Based on this analysis the modified RIVA Score (mRIVA) was developed including age > 60, previous heart surgery, NYHA Class ≥ III, presence of a structural heart disease, epicardial access as well as eGFR as predictive parameters. Logistic regression, the basis for calculating the risk scores, achieved an even better cross-validated accuracy of 73.5± 4.9% based on the mRIVA parameters compared to 70.1± 5.3% for all variables included in the RIVA Score.

Conclusion:  A modified RIVA-Score (mRIVA) that adds patients´ age and heart failure symptoms seems to be a promising option to predict periprocedural outcome after VA ablation and should be validated in further prospective trials.