Risk Stratification in Patients Who Underwent Percutaneous Left Atrial Appendage Occlusion.

Left atrial appendage occlusion (LAAO) is effective in preventing thromboembolism. Risk stratification tools could help identify patients at risk for early mortality after LAAO. In this study, we validated and recalibrated a clinical risk score (CRS) to predict risk of all-cause mortality after LAAO. This study used data from patients who underwent LAAO in a single-center, tertiary hospital. A previously developed CRS using 5 variables (age, body mass index [BMI], diabetes, heart failure, and estimated glomerular filtration rate) was applied to each patient to assess risk of all-cause mortality at 1 and 2 years. The CRS was recalibrated to the present study cohort and compared with established atrial fibrillation-specific (CHA2DS2-VASc and HAS-BLED) and generalized (Walter index) risk scores. Cox proportional hazard models were used to assess the risk of mortality and discrimination was assessed by Harrel C-index. Among 223 patients, the 1- and 2-year mortality rates were 6.7% and 11.2%, respectively. With the original CRS, only low BMI (<23 kg/m2) was a significant predictor of all-cause mortality (hazard ratio [HR] [95% CI] 2.76 [1.03 to 7.35]; p = 0.04). With recalibration, BMI <29 kg/m2 and estimated glomerular filtration rate <60 ml/min/1.73 m2 were significantly associated with an increased risk of death (HR [95% CI] 3.24 [1.29 to 8.13] and 2.48 [1.07 to 5.74], respectively), with a trend toward significance noted for history of heart failure (HR [95% CI] 2.13 [0.97 to 4.67], p = 0.06). Recalibration improved the discriminative ability of the CRS from 0.65 to 0.70 and significantly outperformed established risk scores (CHA2DS2-VASc = 0.58, HAS-BLED = 0.55, Walter index = 0.62). In this single-center, observational study, the recalibrated CRS accurately risk stratified patients who underwent LAAO and significantly outperformed established atrial fibrillation-specific and generalized risk scores. In conclusion, clinical risk scores should be considered as an adjunct to standard of care when evaluating a patient's candidacy for LAAO.

Thirty-Day Readmissions After Upper and Lower Gastrointestinal Hemorrhage: A National Perspective in the United States.

BACKGROUND: Upper gastrointestinal hemorrhage (UGIH) and lower gastrointestinal hemorrhage (LGIH) are 2 of the most common reasons for hospital admissions across the United States. The 30-day readmission after index admission poses a major burden on the health care infrastructure, and thus, it is important to assess the causes of 30-day readmission for patients with UGIH and LGIH. METHODS: The study cohort was derived from the 2013 National Readmission Database. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) Volume 3 diagnosis codes were utilized to identify UGIH and LGIH patients from this data set. Patients who were readmitted to the hospital within 30 days within the same calendar year were further analyzed. Categorical variables and continuous variables were assessed by the χ test and the student t test, respectively. The independent predictors of unplanned 30-day readmissions were recognized by multivariate logistic regression, adjusting for stratified cluster design of National Readmission Database. SAS 9.4 (SAS Institute Inc., Cary, NC) was used for data analysis. RESULTS: The number of index admissions identified from the National Readmission Data 2013 were 82,290 for UGIH and 133,114 for LGIH. All-cause 30-day readmission rate for UGIH versus LGIH was 14.6% (readmitted N=12,046; 56.64% age 65 y and above) versus 14.4% (readmitted N=19,128; 70.21% age 65 y and above and 49.61% men). Gastrointestinal causes were most common (33.9% vs. 39.6%), followed by cardiac (13.3% vs. 15.3%), infectious (10.4% vs. 9.1%), and respiratory causes (7.8% vs. 7.1%) for 30-day readmission for UGIH and LGIH. Significant predictors of increased 30-day readmission (odds ratio, 95% confidence interval, P-value) included metastatic disease (2.15, 1.75-2.64, P<0.001), discharge against medical advice (1.85, 1.55-2.22, P<0.001), and length of stay >3 days (1.50, 1.38-1.63, P<0.001). Predictors for 30-day readmission for LGIH included metastatic disease (1.75, 1.48-2.06, P<0.001), liver disease (1.59, 1.49-1.71, P<0.001), and drug abuse (1.38, 1.21-1.58, P<0.001). CONCLUSIONS: Most common reason for UGIH and LGIH readmission was related to gastrointestinal disease, followed by cardiac, infectious, and respiratory etiologies. By addressing these etiologies for readmission, it may be possible to reduce adverse outcomes.