Abstract
This thesis aimed to explore potential tools for improving cardiovascular (CVD) management, focusing on heart failure (HF) and acute coronary syndrome (ACS).
The first part examined big data applications, ranging from subgroup identification to monitoring. Using retrospective health insurance claims data, risk factors for adverse outcomes in chronic HF were investigated, revealing sex-specific differences in comorbidities but not in medication adherence, further denoting the value of such databases. Innovative machine learning techniques were then deployed, demonstrating their superiority in predictive value for adverse outcomes compared to traditional methods. Additionally, a meta-analysis on home telemonitoring systems (hTMS) showed a significant reduction in adverse outcomes, particularly in non-invasive hTMS studies, advocating their integration into outpatient management. Furthermore, a study protocol for a randomized controlled trial (RCT) aimed to promote physical activity in HF patients was developed. Lastly, PCKS9 inhibitors were found to be well-tolerated in real-world populations, with an adverse events profile comparable to RCTs.
The second part of this thesis focused on risk stratification primarily through the analysis of serial measurements of blood biomarkers in both HF and post-ACS patients. The prognostic value of growth differentiation factor 15 (GDF-15) and other biomarkers was explored. Serial measurements of GDF-15 emerged as a strong predictor of adverse outcomes. Interestingly, concentrations rose before an adverse outcome during follow-up. Additionally, the prognostic value of iron deficiency in post-ACS patients was investigated, highlighting its association with an increased risk for adverse outcomes and its potential as a target in post-ACS management. Lastly, in a heart transplantation database, pre-transplant chronic kidney disease was identified as a significant risk factor for the incidence of malignancy post-transplantation, emphasizing strategies to mitigate these risks pre-transplantation.
Overall, this thesis provides valuable insights into utilizing big data analysis and serial biomarker measurements to enhance clinical decision-making in CVD, specifically focusing on HF and ACS. These findings contribute to advancing personalized medicine approaches that could revolutionize CVD management and mitigate the growing healthcare burden associated with this condition.
The first part examined big data applications, ranging from subgroup identification to monitoring. Using retrospective health insurance claims data, risk factors for adverse outcomes in chronic HF were investigated, revealing sex-specific differences in comorbidities but not in medication adherence, further denoting the value of such databases. Innovative machine learning techniques were then deployed, demonstrating their superiority in predictive value for adverse outcomes compared to traditional methods. Additionally, a meta-analysis on home telemonitoring systems (hTMS) showed a significant reduction in adverse outcomes, particularly in non-invasive hTMS studies, advocating their integration into outpatient management. Furthermore, a study protocol for a randomized controlled trial (RCT) aimed to promote physical activity in HF patients was developed. Lastly, PCKS9 inhibitors were found to be well-tolerated in real-world populations, with an adverse events profile comparable to RCTs.
The second part of this thesis focused on risk stratification primarily through the analysis of serial measurements of blood biomarkers in both HF and post-ACS patients. The prognostic value of growth differentiation factor 15 (GDF-15) and other biomarkers was explored. Serial measurements of GDF-15 emerged as a strong predictor of adverse outcomes. Interestingly, concentrations rose before an adverse outcome during follow-up. Additionally, the prognostic value of iron deficiency in post-ACS patients was investigated, highlighting its association with an increased risk for adverse outcomes and its potential as a target in post-ACS management. Lastly, in a heart transplantation database, pre-transplant chronic kidney disease was identified as a significant risk factor for the incidence of malignancy post-transplantation, emphasizing strategies to mitigate these risks pre-transplantation.
Overall, this thesis provides valuable insights into utilizing big data analysis and serial biomarker measurements to enhance clinical decision-making in CVD, specifically focusing on HF and ACS. These findings contribute to advancing personalized medicine approaches that could revolutionize CVD management and mitigate the growing healthcare burden associated with this condition.
| Original language | English |
|---|---|
| Awarding Institution |
|
| Supervisors/Advisors |
|
| Award date | 20 Mar 2024 |
| Place of Publication | Rotterdam |
| Print ISBNs | 978-94-6483-828-2 |
| Publication status | Published - 20 Mar 2024 |
