Trends in outcomes, cost, and readmissions of transcatheter edge to edge repair in the United States (2014-2018)

Machine learning vs. conventional methods for prediction of 30-day readmission following percutaneous mitral edge-to-edge repair

BACKGROUND

Identifying predictors of readmissions after mitral valve transcatheter edge-to-edge repair (MV-TEER) is essential for risk stratification and optimization of clinical outcomes.

AIMS

We investigated the performance of machine learning [ML] algorithms vs. logistic regression in predicting readmissions after MV-TEER.

METHODS

We utilized the National-Readmission-Database to identify patients who underwent MV-TEER between 2015 and 2018. The database was randomly split into training (70 %) and testing (30 %) sets. Lasso regression was used to remove non-informative variables and rank informative ones. The top 50 informative predictors were tested using 4 ML models: ML-logistic regression [LR], Naive Bayes [NB], random forest [RF], and artificial neural network [ANN]/For comparison, we used a traditional statistical method (principal component analysis logistic regression PCA-LR).

RESULTS

A total of 9425 index hospitalizations for MV-TEER were included. Overall, the 30-day readmission rate was 14.6 %, and heart failure was the most common cause of readmission (32 %). The readmission cohort had a higher burden of comorbidities (median Elixhauser score 5 vs. 3) and frailty score (3.7 vs. 2.9), longer hospital stays (3 vs. 2 days), and higher rates of non-home discharges (17.4 % vs. 8.5 %). The traditional PCA-LR model yielded a modest predictive value (area under the curve [AUC] 0.615 [0.587-0.644]). Two ML algorithms demonstrated superior performance than the traditional PCA-LR model; ML-LR (AUC 0.692 [0.667-0.717]), and NB (AUC 0.724 [0.700-0.748]). RF (AUC 0.62 [0.592-0.677]) and ANN (0.65 [0.623-0.677]) had modest performance.

CONCLUSION

Machine learning algorithms may provide a useful tool for predicting readmissions after MV-TEER using administrative databases.

Temporal Trends and Early Outcomes of Transcatheter versus Surgical Mitral Valve Repair in Atrial Fibrillation Patients

Objectives

To study trends of utilization, in-hospital outcomes, and short outcomes in patients undergoing transcatheter mitral valve repair (TMVR) vs. surgical mitral valve repair (SMVR) in atrial fibrillation (AF).

Background

TMVR is a treatment option in inoperable or high-risk patients with mitral regurgitation (MR). AF is a common comorbidity of MR. Data comparing between TMVR and SMVR in MR patients with AF is lacking.

Methods

The National Readmission Database from 2016 to 2019 was utilized to identify hospitalizations undergoing TMVR or SMVR with AF. Outcomes of interest included mortality, postoperative complications, length of stay, and 30-day readmission rate.

Results

A total of 9,195 patients underwent TMVR and 16,972 patients underwent SMVR with AF; the number of AF undergoing TMVR was increasing from 1,342 in 2016 to 4,215 in 2019 and SMVR. The incidence of in-hospital mortality decreased from 2.6% in 2016 to 1.8% in 2019. We identified length of stay>5 days, dyslipidemia, cerebrovascular disease, heart failure with reduced ejection fraction, and urgent/emergent admissions as independent risk factors for in-hospital mortality. After matching, we included 4,680 patients in each group; the in-hospital death, transfusion, acute kidney injury, sepsis, stroke, and mechanical ventilation were lower in TMVR compared with SMVR. TMVR was associated with a similar rate of all-cause readmission at 30 days compared with SMVR.

Conclusion

Patients with AF receiving TMVR have been increasing along with progressive improvement in in-hospital death and length of stay. Compared to SMVR, AF patients receiving TMVR had a lower rate of in-hospital death and postoperative complications.

Mitral Valve Surgery for Persistent or Recurrent Mitral Regurgitation After Transcatheter Edge-to-Edge Repair Is Associated With Improved Survival

Background

The management of severe mitral regurgitation (MR) after transcatheter edge‐to‐edge repair (TEER) remains a clinical conundrum. Considering the growing volume of TEER, more outcomes data for mitral surgery in this cohort are needed.

Methods and Results

Symptomatic patients with persistent or recurrent severe MR after TEER evaluated between May 2014 and June 2021 were included. The primary outcome was all‐cause mortality in patients who were treated with surgery versus medical therapy. The Kaplan–Meier and Cox regression methods were used to report risk‐adjusted survival analyses. Among the 142 included patients, 44 (31.0%) underwent mitral surgery. Patients who underwent surgery were younger than those treated medically (74.1±8.9 versus 78.6±10.5 years, P=0.01). Major comorbidities were similar except obesity, sleep apnea, left ventricular dimensions, and ejection fraction. Society of Thoracic Surgeons Predicted Risk of Operative Mortality was 9.0±4.7 versus 7.9±4.9 in the surgical versus medical therapy groups, respectively, P=0.22. Time from TEER to detection of severe MR was similar in both groups (median [interquartile range] 97.5 [39.5–384] versus 93.5 [40–389] days in the surgical versus medical groups, respectively [P>0.05]). In the surgical group, valve replacement was performed in all patients. Operative mortality was 4.5% (observed/expected ratio 0.55), and major complications were uncommon. After risk‐adjustment, surgery was associated with significantly lower all‐cause mortality (adjusted hazard ratio, 0.33 [95% CI, 0.12–0.92], P=0.001) compared with medial therapy.

Conclusions

Compared with medical therapy, mitral surgery in patients with severe persistent or recurrent MR after TEER is associated with lower mortality despite the high‐risk profile of these patients. Patients with severe MR after TEER should be considered for surgery at a referral mitral surgical center.

Gender-related differences in clinical outcomes after either single or double left main bifurcation stenting

We sought to examine the impact of gender differences in clinical outcomes at 3 years also comparing the role of double versus single stenting approach for the treatment of coronary unprotected LM bifurcation lesions. We retrospectively analyzed both the procedural and medical data of patients referred to our hub center for complex LM bifurcation disease, treated using Crossover provisional stenting, T or T-and-Protrusion (TAP), Culotte, and Nano-inverted-T (NIT) techniques between January 1st, 2008 and May 1st 2018. The main outcome of the study was to evaluate the association between gender and target lesion failure (TLF) based on the different stenting technique used. Five hundred and sixty-seven patients (251 females, mean age 70.0 ± 10 years, mean Syntax score 31.6 ± 6.3) were evaluated. Crossover, T or TAP, culotte and NIT techniques were performed in 171 (30.1%), 61 (10.7%), 98 (17.2%) and 237 (41.8%) patients, respectively with no differences in baseline and peri-procedural items among gender. At a mean follow-up of 37.1 ± 10.8 months (range 22.1-39.3 moths), the overall TLF rate, cardiovascular mortality and stent thrombosis were 12.1%, 3.1% and 1.0%, respectively. Female gender was associated with an increased rate of major bleeding when treated with double stent strategy (p = 0.02). No gender difference in TLF was noted among gender, independently from the stenting approach used. Among patients with ULM bifurcation disease undergoing PCI, TLF rates were not different between genders at 3-year follow-up either using a single or double stent technique.