Recursive partitioning-based preoperative risk stratification for atrial fibrillation after coronary artery bypass surgery

Incidence and risk factors for new-onset atrial fibrillation following coronary artery bypass grafting: A systematic review and meta-analysis

OBJECTIVES

To estimate the incidence of new-onset post-operative atrial fibrillation after isolated coronary artery bypass surgery and summarise the evidence on risk factors that predispose people to developing the complication.

STUDY DESIGN/METHODS

A systematic review was conducted to identify studies from the CINAHL, MEDLINE and Cochrane databases. A title and abstract review was conducted by one reviewer. Full text review and quality assessment processes were conducted by two reviewers. Incidence data was combined in meta-analysis using the 'metaprop' routine in Stata and risk factor data were synthesised in narrative and table format.

RESULTS

Ten studies, including 6173 participants, were included in the review. The estimated pooled incidence of post-operative atrial fibrillation was 25% (CI 0.19-0.30). In a secondary meta-analysis including studies that only included first time bypass surgery recipients the estimated pooled incidence was 26% (CI 0.14-0.41). Due to high levels of heterogeneity these results should be interpreted with caution. Risk factors with the strongest associations to post-operative atrial fibrillation were chronic obstructive pulmonary disease, decreased partial pressure of oxygen on air, congestive heart failure, right coronary artery disease, male gender, prolonged cross clamp time and port-operative inotropic exposure.

CONCLUSION

Further prospective studies are needed to strengthen the current evidence base.

A systematic review of the clinical application of data-driven population segmentation analysis

BACKGROUND

Data-driven population segmentation analysis utilizes data analytics to divide a heterogeneous population into parsimonious and relatively homogenous groups with similar healthcare characteristics. It is a promising patient-centric analysis that enables effective integrated healthcare interventions specific for each segment. Although widely applied, there is no systematic review on the clinical application of data-driven population segmentation analysis.

METHODS

We carried out a systematic literature search using PubMed, Embase and Web of Science following PRISMA criteria. We included English peer-reviewed articles that applied data-driven population segmentation analysis on empirical health data. We summarized the clinical settings in which segmentation analysis was applied, compared and contrasted strengths, limitations, and practical considerations of different segmentation methods, and assessed the segmentation outcome of all included studies. The studies were assessed by two independent reviewers.

RESULTS

We retrieved 14,514 articles and included 216 articles. Data-driven population segmentation analysis was widely used in different clinical contexts. 163 studies examined the general population while 53 focused on specific population with certain diseases or conditions, including psychological, oncological, respiratory, cardiovascular, and gastrointestinal conditions. Variables used for segmentation in the studies are heterogeneous. Most studies (n = 170) utilized secondary data in community settings (n = 185). The most common segmentation method was latent class/profile/transition/growth analysis (n = 96) followed by K-means cluster analysis (n = 60) and hierarchical analysis (n = 50), each having its advantages, disadvantages, and practical considerations. We also identified key criteria to evaluate a segmentation framework: internal validity, external validity, identifiability/interpretability, substantiality, stability, actionability/accessibility, and parsimony.

CONCLUSIONS

Data-driven population segmentation has been widely applied and holds great potential in managing population health. The evaluations of segmentation outcome require the interplay of data analytics and subject matter expertise. The optimal framework for segmentation requires further research.

Development of a Nonparametric Predictive Model for Readmission Risk in Elderly Adults After Colon and Rectal Cancer Surgery

OBJECTIVES

Primary objective: to use advanced nonparametric techniques to determine risk factors for readmission after colorectal cancer surgery in elderly adults.

SECONDARY OBJECTIVE

to compare this methodology with traditional parametric methods.

DESIGN

Using data from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP), nonparametric techniques were used to evaluate the risk of readmission in elderly adults undergoing surgery for colorectal cancer in 2011 and 2012.

SETTING

More than 200 hospitals participating in the NSQIP database.

PARTICIPANTS

Individuals aged 65 and older who underwent surgery for colorectal cancer in 2011 and 2012 (N = 2,117).

MEASUREMENTS

Age-stratified robust nonparametric predictive model (classification and regression tree (CART) analysis) of 30-day readmission for elderly adults undergoing surgery for colorectal cancer.

RESULTS

Recent chemotherapy was the most important predictor of readmission in participants aged 65 to 74, with 20% of those with recent chemotherapy and 11% of with no recent chemotherapy being readmitted. Participants aged 75 to 84 who had recently undergone chemotherapy had a readmission rate of 23%, whereas those with no chemotherapy had a readmission rate of 9%. Being underweight was the greatest predictor of readmission (30%) in participants aged 85 and older. These methods were found to be more robust than traditional logistic regression.

CONCLUSION

Specific age-related preoperative factors help predict readmission in elderly adults undergoing colorectal cancer surgery. Results of the nonparametric CART analysis are better than traditional regression analysis and help physicians to clinically stratify based on age. This model may help identify individuals in whom intervention may be helpful in reducing readmission after surgery.

Response shift in patients with multiple sclerosis: an application of three statistical techniques

OBJECTIVE

With the evolution of theory and methods for detecting recalibration, reprioritization, and reconceptualization response shifts, the time has come to evaluate and compare the current statistical detection techniques. This manuscript presents an overview of a cross-method validation done on the same patient sample.

METHODS

Three statistical techniques were used: Structural Equation Modeling, Latent Trajectory Analysis, and Recursive Partitioning and Regression Tree modeling. The study sample (n = 3,008) was drawn from the North American Research Committee on Multiple Sclerosis (NARCOMS) Registry to represent patients soon after diagnosis, classified as having either a self-reported relapsing, progressive, or stable disease trajectory. Patient-reported outcomes included the disease-specific Performance Scales and the Patient-Derived Disease Steps, and the generic SF-12v2 measure.

RESULTS

Small response shift effect sizes were detected by all of the methods. Recalibration response shift was detected by Structural Equation Modeling, Recursive Partitioning Regression Tree demonstrated patterns consistent with all three types of response shift, and Latent Trajectory Analysis, although unable to distinguish types of response shift, did detect response shift in less than 1% of the sample.

CONCLUSION

The methods and their findings were discussed for operationalization, interpretability, assumptions, ability to use all data points from the study sample, limitations, and strengths. Directions for future research are discussed.

Predictors of atrial fibrillation following coronary artery bypass surgery

BACKGROUND

New-onset atrial fibrillation is the most common form of rhythm disturbance following coronary artery bypass grafting surgery (CABG). It is still unclear which factors have a significant impact on its occurrence after this procedure. The aim of this study was to evaluate clinical predictors of postoperative atrial fibrillation (POAF) after myocardial revascularization.

MATERIAL/METHODS

We performed a retrospective analysis of 322 patients who underwent the first CABG operation without baseline atrial fibrillation. All subjects underwent laboratory blood tests, echocardiography and selective coronarography with ventriculography. Patients were continuously electrocardiographically monitored during the first 48-72 h after the operation for the occurrence of POAF.

RESULTS

POAF was diagnosed in 72 (22.4%) of the patients. Multivariate logistic regression analysis was used to identify the following independent clinical predictors of POAF: age≥65 years (OR 1.78; 95%CI: 1.06-2.76; p=0.043), hypertension (OR 1.97; 95%CI: 1.15-3.21; p=0.018), diabetes mellitus (OR 2.09; 95% CI: 1.31-5.33; p=0.010), obesity (OR 1.51; 95%CI: 1.03-3.87; p=0.031), hypercholesterolemia (OR 2.17, 95%CI: 1.05-4.25; p=0.027), leukocytosis (OR 2.32, 95%CI: 1.45-5.24; p=0.037), and left ventricular segmental kinetic disturbances (OR 3.01; 95%CI: 1.65-4.61, p<0.001).

CONCLUSIONS

This study demonstrates that advanced age, hypertension, diabetes, obesity, hypercholesterolemia, leukocytosis, and segmental kinetic disturbances of the left ventricle are powerful risk factors for the occurrence of POAF.

Pretreatment diffusion- and perfusion-MR lesion volumes have a crucial influence on clinical response to stroke thrombolysis

We hypothesized that pretreatment magnetic resonance imaging (MRI) diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) lesion volumes may have influenced clinical response to thrombolysis in the Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET). In 98 patients randomized to intravenous (IV) tissue plasminogen activator (tPA) or placebo 3 to 6 h after stroke onset, we examined increasing acute DWI and PWI lesion volumes (Tmax-with 2-sec delay increments), and increasing PWI/DWI mismatch ratios, on the odds of both excellent (modified Rankin Scale (mRS): 0 to 1) and poor (mRS: 5 to 6) clinical outcome. Patients with very large PWI lesions (most had internal carotid artery occlusion) had increased odds ratio (OR) of poor outcome with IV-tPA (58% versus 25% placebo; OR=4.13, P=0.032 for Tmax +2-sec volume >190 mL). Excellent outcome from tPA treatment was substantially increased in patients with DWI lesions <18 mL (77% versus 18% placebo, OR=15.0, P<0.001). Benefit from tPA was also seen with DWI lesions up to 25 mL (69% versus 29% placebo, OR=5.5, P=0.03), but not for DWI lesions >25 mL. In contrast, increasing mismatch ratios did not influence the odds of excellent outcome with tPA. Clinical responsiveness to IV-tPA, and stroke outcome, depends more on baseline DWI and PWI lesion volumes than the extent of perfusion-diffusion mismatch.

Classification and regression tree uncovered hierarchy of psychosocial determinants underlying quality-of-life response shift in HIV/AIDS

OBJECTIVES

Rapkin and Schwartz define response shift as otherwise unexplained, discrepant change in health-related quality of life (HRQOL) that is associated with change in cognitive appraisal. In this article, we demonstrate how a recursive partitioning (rpart) regression tree analytic approach may be used to explore cognitive changes to gain additional insight into response-shift phenomena.

STUDY DESIGN AND SETTING

Data are from the "Choices in Care Study," an evaluation of HIV+ Medicaid recipients' experiences and outcomes in care (N=394). Cognitive assessment was based on the QOL appraisal battery. HRQOL was measured by the SF-36 Health Survey, version 2 (SF-36v2).

RESULTS

We used rpart to examine 6-month change in SF-36v2 mental composite score as a function of changes in appraisal, after controlling for patient characteristics, health changes, and intervening events. Rpart identified nine distinct patterns of cognitive change, including three associated with negative discrepancies, four with positive discrepancies, and two with no discrepancies.

CONCLUSION

Rpart classification provides a nuanced treatment of response shift. This methodology has implications for evaluating programs, guiding decisions, and targeting care.

Predictors of prolonged postoperative endotracheal intubation in patients undergoing thoracotomy for lung resection

OBJECTIVE

The aim of this study was to identify predictors of delayed endotracheal extubation defined as the need for postoperative ventilatory support after open thoracotomy for lung resection.

DESIGN

An observational cohort investigation.

SETTING

A tertiary referral center.

PARTICIPANTS

The study population consisted of 2,068 patients who had open thoracotomy for pneumonectomy, lobectomy, or segmental lung resection between January 1996 and December 2005.

INTERVENTIONS

Not applicable.

MEASUREMENTS AND MAIN RESULTS

Preoperative and intraoperative variables were collected concurrently with the patient's care. Risk factors were identified using logistic regression with stepwise variable selection procedure on 1,000 bootstrap resamples, and a bagging algorithm was used to summarize the results. Intraoperative red blood cell transfusion, higher preoperative serum creatinine level, absence of a thoracic epidural catheter, more extensive surgical resection, and lower preoperative FEV(1) were associated with an increased risk of delayed extubation after lung resection.

CONCLUSION

Most predictors of delayed postoperative extubation (ie, red blood cell transfusion, higher preoperative serum creatinine, lower preoperative FEV(1), and more extensive lung resection) are difficult to modify in the perioperative period and probably represent greater severity of underlying lung disease and more advanced comorbid conditions. However, thoracic epidural anesthesia and analgesia is a modifiable factor that was associated with reduced odds for postoperative ventilatory support. Thus, the use of epidural analgesia may reduce the need for post-thoracotomy mechanical ventilation.