Anxiety, Cardiovascular Diseases, Coronary Artery Disease, Depression, Depressive Symptoms, Heart Diseases, Postoperative Complications, Posttraumatic Stress Disorder, Psychological Well-Being, Quality of Life
Conditions
Keywords
Coronary Artery Bypass Graft, Coronary Artery Bypass Grafting, Coronary Artery Bypass Surgery, CABG, Inflammation, Interoception, Clinically Significant Depressive Symptoms, CSDS, Patient Health Questionnaire-9, PHQ-9, Postoperative Depression, Algorithms, Prediction, Predictive Modeling, Machine Learning
Brief summary
The primary goal of this project is to develop a predictive model for clinically significant depressive symptoms (CSDS) in patients undergoing coronary artery bypass graft (CABG) surgery, using pre- and perioperative data. CSDS occur in about 30 percent of CABG patients, which is four times higher than in the general population. These symptoms are linked to poor quality of life and increased morbidity and mortality. The aim is to create a model that can identify patients at risk for postoperative depression. This tool could help clinicians make informed decisions and take preventive measures to manage depression after surgery.
Detailed description
In patients undergoing coronary artery bypass graft (CABG) surgery, the prevalence of clinically significant depressive symptoms (CSDS) is about 30 percent, four times higher than the 12-month prevalence in the general population. CSDS are associated with poor quality of life and increased morbidity and mortality. While several predictors of post-CABG CSDS have been identified, no prognostic model exists. The aim of this project is to develop a predictive model for post-surgery CSDS in CABG patients using pre- and perioperative data. A prognostic prediction model for CSDS 6 weeks post-CABG, will be developed using demographic, psychometric, medical, inflammation, and cardiac interoception data. Machine learning algorithms will be employed for data analysis. A cohort of 350 participants from two hospitals will be recruited, with 300 participants expected to complete the study. Data will be divided into training (200 participants) and testing (100 participants) sets. Nested cross-validation will prevent overfitting. Both binary and regression prediction models will be used. Additionally, a simpler model will be developed to increase generalizability. The prediction model will identify CABG patients at risk of post-surgery CSDS. The model will help identify patients at risk for CSDS before surgery, enabling early interventions. Clinicians can make precision medicine decisions to prevent or manage CSDS, improving postoperative psychological well-being. Additionally, the study could advance understanding of the mechanisms linking depression and coronary heart disease, particularly in relation to inflammation and interoception.
Interventions
None listed
Sponsors
Triemli Hospital
ETH Zurich
University of Zurich
Roland von Känel
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to 90 Years
Inclusion criteria
* Elective Off-Pump CABG or CABG, either isolated or combined with valve intervention * Men and women, aged between 18 and 90 years * Sufficient knowledge of German language in reading and understanding * Oral and Signed consent form * Ability and Willingness to follow the study protocol
Exclusion criteria
* Cognitive impairment according to a score of ≤ 7 (maximum score = 9) on a modified version of a short version of the Mini-Mental State Examination and the Brief Interview for Mental Status * Any serious comorbid non-cardiac medical condition likely to cause death within 1 year (e.g. metastatic cancer) * Active psychotic symptoms (assessed with two items from the Youth Psychosis At Risk Questionnaire), substance abuse and/or dependence within the past 6 months (assessed with a single-item questionnaire), and/or active suicidal ideations (assessed with a single item from the M.I.N.I.). * Acute or emergency CABG
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Patient Health Questionnaire (PHQ-9) score ≥10 (yes/no) at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The Patient Health Questionnaire (PHQ)-9 will assess the severity of self-rated depressive symptoms over the last two weeks. It covers the nine Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5) criteria for major depression, with symptoms rated on a 4-point Likert scale. Scores range from 0 to 27, with higher scores indicating more severe symptoms. A score of 10 or higher corresponds to a diagnosis of depression, with 88 percent sensitivity and specificity. As the best cut-off for post-CABG CSDS is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. One approach frames the prediction challenge as a binary classification problem and uses a PHQ-9 cut-off score ≥10 for defining the presence versus absence of CSDS. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| General Anxiety Disorder (GAD-7) score ≥10 (yes/no) at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The General Anxiety Disorder (GAD)-7 questionnaire will assess self-rated anxiety symptoms over the past two weeks. Seven items are rated on a 4-point Likert scale, with total scores ranging from 0 to 21. A score of 10 or higher indicates moderate to severe anxiety, corresponding to a GAD diagnosis with 89% sensitivity and 82% specificity. As the best cut-off for post-CABG GAD is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. One approach frames the prediction challenge as a binary classification problem and uses a GAD-7 cut-off score ≥10 for defining the presence versus absence of GAD. |
| GAD-7 score continuous at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The GAD-7 questionnaire will assess self-rated anxiety symptoms over the past two weeks. Seven items are rated on a 4-point Likert scale, with total scores ranging from 0 to 21. A score of 10 or higher indicates moderate to severe anxiety, corresponding to a GAD diagnosis with 89% sensitivity and 82% specificity. As the best cut-off for post-CABG GAD is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. The second approach views the prediction challenge as a regression problem and tries to predict individual GAD-7 scores without applying any threshold. |
| PTSD (Post-traumatic Stress Disorder) Checklist for DSM-5 (PCL-5) score ≥33 at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The PTSD Checklist for DSM-5 (PCL-5) will assess CABG surgery-induced posttraumatic stress. It is a 20-item self-report measure evaluating the DSM-5 PTSD symptoms over the past month. Items are rated on a 5-point Likert scale, with scores of 33 or higher indicating probable PTSD. As the best cut-off for post-CABG PTSD is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. One approach frames the prediction challenge as a binary classification problem and uses a PCL-5 cut-off score ≥33 for defining the presence versus absence of PTSD. |
| PHQ-9 score continuous at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The PHQ-9 will assess the severity of self-rated depressive symptoms over the last two weeks. It covers the nine DSM-5 criteria for major depression, with symptoms rated on a 4-point Likert scale. Scores range from 0 to 27, with higher scores indicating more severe symptoms. A score of 10 or higher corresponds to a diagnosis of depression, with 88 percent sensitivity and specificity. As the best cut-off for post-CABG CSDS is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. The second approach views the prediction challenge as a regression problem and tries to predict individual PHQ-9 scores without applying any threshold. |
| Short-Form Health Survey-12 (SF-12) mental health component score at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The Short-Form Health Survey (SF-12) will assess physical and mental health-related quality of life (QoL) over the last four weeks, covering aspects such as physical functioning, pain, vitality, social functioning, and mental health. |
| SF-12 physical health component score at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The SF-12 will assess physical and mental health-related quality of life (QoL) over the last four weeks, covering aspects such as physical functioning, pain, vitality, social functioning, and mental health. |
| PCL-5 score continuous at 6 weeks post-CABG | Baseline (1 day before CABG surgery) and Follow Up (6 weeks after CABG surgery) | The PCL-5 will assess CABG surgery-induced posttraumatic stress. It is a 20-item self-report measure evaluating the DSM-5 PTSD symptoms over the past month. Items are rated on a 5-point Likert scale, with scores of 33 or higher indicating probable PTSD. As the best cut-off for post-CABG PTSD is difficult to determine a priori for a prediction model, a two complementary approaches for the analysis will be used, correcting for multiple tests when assessing the significance of the accuracy of the prediction model. The second approach views the prediction challenge as a regression problem and tries to predict individual PCL-5 scores without applying any threshold. |
Countries
Switzerland
Contacts
Primary ContactRoland v Känel, Prof. Dr.
Backup ContactSinthujan Sivakumar, MSc
Outcome results
None listed