| 1. |
- Adler, Eric D., et al.
(author)
-
Improving risk prediction in heart failure using machine learning
- 2020
-
In: European Journal of Heart Failure. - : Wiley. - 1388-9842 .- 1879-0844. ; 22:1, s. 139-147
-
Journal article (peer-reviewed)abstract
- Background: Predicting mortality is important in patients with heart failure (HF). However, current strategies for predicting risk are only modestly successful, likely because they are derived from statistical analysis methods that fail to capture prognostic information in large data sets containing multi-dimensional interactions. Methods and results: We used a machine learning algorithm to capture correlations between patient characteristics and mortality. A model was built by training a boosted decision tree algorithm to relate a subset of the patient data with a very high or very low mortality risk in a cohort of 5822 hospitalized and ambulatory patients with HF. From this model we derived a risk score that accurately discriminated between low and high-risk of death by identifying eight variables (diastolic blood pressure, creatinine, blood urea nitrogen, haemoglobin, white blood cell count, platelets, albumin, and red blood cell distribution width). This risk score had an area under the curve (AUC) of 0.88 and was predictive across the full spectrum of risk. External validation in two separate HF populations gave AUCs of 0.84 and 0.81, which were superior to those obtained with two available risk scores in these same populations. Conclusions: Using machine learning and readily available variables, we generated and validated a mortality risk score in patients with HF that was more accurate than other risk scores to which it was compared. These results support the use of this machine learning approach for the evaluation of patients with HF and in other settings where predicting risk has been challenging.
|
|
| 2. |
- Ammirati, Enrico, et al.
(author)
-
Outcome of patients on heart transplant list treated with a continuous-flow left ventricular assist device : Insights from the TRans-Atlantic registry on VAd and TrAnsplant (TRAViATA)
- 2021
-
In: International Journal of Cardiology. - : Elsevier BV. - 0167-5273. ; 324
-
Journal article (peer-reviewed)abstract
- Background: Geographic variations in management and outcomes of individuals supported by continuous-flow left ventricular assist devices (CF-LVAD) between the United States (US) and Europe (EU) is largely unknown. Methods: We created a retrospective, multinational registry of 524 patients who received a CF-LVAD (either HVAD or Heartmate II) between January 2008 and April 2017. Follow up spanned from date of CF-LVAD implant to post-HTx period with a median follow up of 44.8 months. Results: The cohort included 299 (57.1%) EU and 225 (42.9%) US patients. Although the US cohort was significantly older with a higher prevalence of comorbidities, survival was similar between the cohorts (US 63.1%, EU 68.4% at 5 years, unadjusted log-rank test p = 0.43).Multivariate analyses suggested that older age, higher body mass index, elevated creatinine, use of temporary mechanical circulatory support prior CF-LVAD, and implantation of HVAD were associated with increased mortality. Among CF-LVAD patients undergoing HTx, the median time on CF-LVAD support was shorter in the US, meanwhile US donors were younger. Finally, the pattern of adverse events (stroke, gastrointestinal bleedings, late right ventricular failure, and driveline infection) during support differed significantly between US and EU. Conclusions: Although waitlisted patients in the US on CF-LVAD have higher risk comorbid conditions, the overall outcome is similar in US and EU. Geographic variations with regards to donor characteristics, duration of CF-LVAD support prior to transplant, and adverse events on support can explain the disparity in the utilization of mechanical bridge to transplant strategy between US and EU.
|
|
| 3. |
- Darden, Douglas, et al.
(author)
-
Cardiovascular implantable electronic device therapy in patients with left ventricular assist devices : insights from TRAViATA
- 2021
-
In: International Journal of Cardiology. - : Elsevier BV. - 0167-5273. ; 340, s. 26-33
-
Journal article (peer-reviewed)abstract
- Background: There is conflicting observational data on the survival benefit cardiac implantable electronic devices (CIED) in patients with LVADs. Methods: Patients in whom an LVAD was implanted between January 2008 and April 2017 in the multinational Trans-Atlantic Registry on VAD and Transplant (TRAViATA) registry were separated into four groups based on the presence of CIED prior to LVAD implantation: none (n = 146), implantable cardiac defibrillator (ICD) (n = 239), cardiac resynchronization without defibrillator (CRT-P) (n = 28), and CRT with defibrillator (CRT-D) (n = 111). Results: A total of 524 patients (age 52 years ±12, 84.4% male) were followed for 354 (interquartile range: 166–701) days. After multivariable adjustment, there were no differences in survival across the groups. In comparison to no device, only CRT-D was associated with late right ventricular failure (RVF) (hazard ratio 2.85, 95% confidence interval [CI] 1.42–5.72, p = 0.003). There was no difference in risk of early RVF across the groups or risk of ICD shocks between those with ICD and CRT-D. Conclusion: In a multinational registry of patients with LVADs, there were no differences in survival with respect to CIED subtype. However, patients with a pre-existing CRT-D had a higher likelihood of late RVF suggesting significant long-term morbidity in those with devices capable of LV‑lead pacing post LVAD implantation.
|
|
| 4. |
- Kalantarian, Shadi, et al.
(author)
-
Long-Term Electrocardiographic and Echocardiographic Progression of Arrhythmogenic Right Ventricular Cardiomyopathy and Their Correlation With Ventricular Tachyarrhythmias.
- 2021
-
In: Circulation Heart Failure. - : LIPPINCOTT WILLIAMS & WILKINS. - 1941-3289 .- 1941-3297. ; 14:9
-
Journal article (peer-reviewed)abstract
- BACKGROUND: Prior studies of structural and electrocardiographic changes in arrhythmogenic right ventricular (RV) cardiomyopathy and their role in predicting ventricular arrhythmias (ventricular tachycardia) have shown conflicting results.METHODS: We reviewed 405 ECGs, 315 transthoracic echocardiographies, and 441 implantable cardioverter defibrillator interrogations in 64 arrhythmogenic RV cardiomyopathy patients (56% men, mean age [SD], 44.2 [14.6] years) over a mean follow-up of 10 (range, 2.3-19) years. Generalized estimating equations were used to identify the association between ECG abnormalities, clinical variables, and transthoracic echocardiographic measurements (>mild degree of tricuspid regurgitation, RV outflow tract diameter in parasternal long axis and short axis, RV end-diastolic area, fractional area change).RESULTS: There was a 4.65 (95% CI, 0.51%-8.8%) increase in RV end-diastolic area, a 3.75 (95% CI, 1.17%-6.34%) decrease in fractional area change, and 1.9 (95% CI, 1.3-2.8) higher odds (odds ratio) of RV wall motion abnormality with every 5-year increase in age after patients' first transthoracic echocardiography. >Mild tricuspid regurgitation was an independent predictor of RV enlargement and dysfunction (hazard ratio of >10% drop in fractional area change from baseline [95% CI], 3.51 [1.77-6.95] and hazard ratio of >10% increase in RV end-diastolic area from baseline [95% CI], 4.90 [2.52-9.52]). Patients with implantable cardioverter defibrillator were more likely to develop >mild tricuspid regurgitation and larger structural and functional disease progression. More pronounced increase in RV end-diastolic area was translated into higher rates of any ventricular tachycardia. Inferior T-wave inversions and sum of R waves (mm) in V1 to V3 were predictors of RV enlargement and dysfunction with the former also predicting risk of any ventricular tachycardia.CONCLUSIONS: Arrhythmogenic RV cardiomyopathy is a progressive disease. Tricuspid regurgitation is an independent predictor of structural disease progression, which may be exacerbated by use of a transvenous implantable cardioverter defibrillator lead.
|
|
