Session Type: ePosters
Session Title: ePosters
Authors(s): Y. Porat (1, 2), J. Nashashibi (1), M. Paul (1)
Authors Affiliations(s): (1) Rambam Health Care Campus, Israel, (2) Technion – Israel Institute of Technology, Israel
Background:
Sepsis carries a high mortality rate and among survivors, functional-cognitive decline and re-infections are common. Hospital readmission is a marker of the severe morbidity following sepsis. We examined risk factors for 30-day readmission among adult patients surviving hospitalization with Gram-negative bacteremia.
Methods:Single-center retrospective cohort study in Israel, including adults (>18 years) experiencing a clinically-significant Gram-negative bacteremia during hospitalization in a medical ward between 2011-2019. We collected a broad data set of variables representing the patients' demographics, comorbidities, the index (first) hospitalization and the bacteremia, including sepsis presentation, severity, the pathogen and its susceptibilities, time to appropriate antibiotic treatment and treatment duration. Data were retrieved from electronic patient charts using a computerized data mining tool. We hypothesized that cephalosporin or carbapenem resistance will be associated with a higher readmission rate. To account of the competing event of deaths between discharge and readmission, we conducted a multivariable competing-risk Cox regression model with 30-days readmission (requiring hospitalization) being the event of interest and deaths the competing event.
Results:Of 3,222 patients with Gram-negative bacteremia, 2,196 survived to discharge and 432 (19.6%) were readmitted within 30 days of discharge. Mean patients' age was 70+/-16 years. The bacteremia was hospital-acquired in 1,082 (49.3%) and healthcare-associated in 168 (7.7%). The most common pathogens isolated in blood were E. Coli, (997, 45.4% of patients). Cephalosporin or carbapenem resistance was observed among 119/338 (35.2%) of readmitted patients compared to 352/1,498 (23.5%) patients without readmission. The covariates retained in the final model predicting 30-day readmissions included nationality, active malignancy, pre-discharge heart rate, discharge with a nasogastric tube, immunosuppressive therapy, anticoagulation therapy, anxiolytic therapy, cephalosporin resistance, and total duration of antibiotic therapy. Appropriate empirical antibiotic treatment was not associated with survivors' readmissions (Table). We plan to present validation of the model in a different hospital.
Conclusions:Risk factors for hospital readmission in the month following sepsis include mainly baseline comorbidities and functional characteristics. However, ESBL phenotypes also contributed to the post-bacteremia morbidity. The risk factors identified can be used for assessing a patient’s readiness for discharge and to target post-discharge interventions for those discharged.
Keyword(s): Gran-negative bacteremia, Readmission, Multidrug resistanceSession Type: ePosters
Session Title: ePosters
Authors(s): Y. Porat (1, 2), J. Nashashibi (1), M. Paul (1)
Authors Affiliations(s): (1) Rambam Health Care Campus, Israel, (2) Technion – Israel Institute of Technology, Israel
Background:
Sepsis carries a high mortality rate and among survivors, functional-cognitive decline and re-infections are common. Hospital readmission is a marker of the severe morbidity following sepsis. We examined risk factors for 30-day readmission among adult patients surviving hospitalization with Gram-negative bacteremia.
Methods:Single-center retrospective cohort study in Israel, including adults (>18 years) experiencing a clinically-significant Gram-negative bacteremia during hospitalization in a medical ward between 2011-2019. We collected a broad data set of variables representing the patients' demographics, comorbidities, the index (first) hospitalization and the bacteremia, including sepsis presentation, severity, the pathogen and its susceptibilities, time to appropriate antibiotic treatment and treatment duration. Data were retrieved from electronic patient charts using a computerized data mining tool. We hypothesized that cephalosporin or carbapenem resistance will be associated with a higher readmission rate. To account of the competing event of deaths between discharge and readmission, we conducted a multivariable competing-risk Cox regression model with 30-days readmission (requiring hospitalization) being the event of interest and deaths the competing event.
Results:Of 3,222 patients with Gram-negative bacteremia, 2,196 survived to discharge and 432 (19.6%) were readmitted within 30 days of discharge. Mean patients' age was 70+/-16 years. The bacteremia was hospital-acquired in 1,082 (49.3%) and healthcare-associated in 168 (7.7%). The most common pathogens isolated in blood were E. Coli, (997, 45.4% of patients). Cephalosporin or carbapenem resistance was observed among 119/338 (35.2%) of readmitted patients compared to 352/1,498 (23.5%) patients without readmission. The covariates retained in the final model predicting 30-day readmissions included nationality, active malignancy, pre-discharge heart rate, discharge with a nasogastric tube, immunosuppressive therapy, anticoagulation therapy, anxiolytic therapy, cephalosporin resistance, and total duration of antibiotic therapy. Appropriate empirical antibiotic treatment was not associated with survivors' readmissions (Table). We plan to present validation of the model in a different hospital.
Conclusions:Risk factors for hospital readmission in the month following sepsis include mainly baseline comorbidities and functional characteristics. However, ESBL phenotypes also contributed to the post-bacteremia morbidity. The risk factors identified can be used for assessing a patient’s readiness for discharge and to target post-discharge interventions for those discharged.
Keyword(s): Gran-negative bacteremia, Readmission, Multidrug resistance
Technion – Israel Institute of Technology, Haifa ,Israel
Session Type: ePosters
Session Title: ePosters
Authors(s): Y. Porat (1, 2), J. Nashashibi (1), M. Paul (1)
Authors Affiliations(s): (1) Rambam Health Care Campus, Israel, (2) Technion – Israel Institute of Technology, Israel
Background:
Sepsis carries a high mortality rate and among survivors, functional-cognitive decline and re-infections are common. Hospital readmission is a marker of the severe morbidity following sepsis. We examined risk factors for 30-day readmission among adult patients surviving hospitalization with Gram-negative bacteremia.
Methods:Single-center retrospective cohort study in Israel, including adults (>18 years) experiencing a clinically-significant Gram-negative bacteremia during hospitalization in a medical ward between 2011-2019. We collected a broad data set of variables representing the patients' demographics, comorbidities, the index (first) hospitalization and the bacteremia, including sepsis presentation, severity, the pathogen and its susceptibilities, time to appropriate antibiotic treatment and treatment duration. Data were retrieved from electronic patient charts using a computerized data mining tool. We hypothesized that cephalosporin or carbapenem resistance will be associated with a higher readmission rate. To account of the competing event of deaths between discharge and readmission, we conducted a multivariable competing-risk Cox regression model with 30-days readmission (requiring hospitalization) being the event of interest and deaths the competing event.
Results:Of 3,222 patients with Gram-negative bacteremia, 2,196 survived to discharge and 432 (19.6%) were readmitted within 30 days of discharge. Mean patients' age was 70+/-16 years. The bacteremia was hospital-acquired in 1,082 (49.3%) and healthcare-associated in 168 (7.7%). The most common pathogens isolated in blood were E. Coli, (997, 45.4% of patients). Cephalosporin or carbapenem resistance was observed among 119/338 (35.2%) of readmitted patients compared to 352/1,498 (23.5%) patients without readmission. The covariates retained in the final model predicting 30-day readmissions included nationality, active malignancy, pre-discharge heart rate, discharge with a nasogastric tube, immunosuppressive therapy, anticoagulation therapy, anxiolytic therapy, cephalosporin resistance, and total duration of antibiotic therapy. Appropriate empirical antibiotic treatment was not associated with survivors' readmissions (Table). We plan to present validation of the model in a different hospital.
Conclusions:Risk factors for hospital readmission in the month following sepsis include mainly baseline comorbidities and functional characteristics. However, ESBL phenotypes also contributed to the post-bacteremia morbidity. The risk factors identified can be used for assessing a patient’s readiness for discharge and to target post-discharge interventions for those discharged.
Keyword(s): Gran-negative bacteremia, Readmission, Multidrug resistanceSession Type: ePosters
Session Title: ePosters
Authors(s): Y. Porat (1, 2), J. Nashashibi (1), M. Paul (1)
Authors Affiliations(s): (1) Rambam Health Care Campus, Israel, (2) Technion – Israel Institute of Technology, Israel
Background:
Sepsis carries a high mortality rate and among survivors, functional-cognitive decline and re-infections are common. Hospital readmission is a marker of the severe morbidity following sepsis. We examined risk factors for 30-day readmission among adult patients surviving hospitalization with Gram-negative bacteremia.
Methods:Single-center retrospective cohort study in Israel, including adults (>18 years) experiencing a clinically-significant Gram-negative bacteremia during hospitalization in a medical ward between 2011-2019. We collected a broad data set of variables representing the patients' demographics, comorbidities, the index (first) hospitalization and the bacteremia, including sepsis presentation, severity, the pathogen and its susceptibilities, time to appropriate antibiotic treatment and treatment duration. Data were retrieved from electronic patient charts using a computerized data mining tool. We hypothesized that cephalosporin or carbapenem resistance will be associated with a higher readmission rate. To account of the competing event of deaths between discharge and readmission, we conducted a multivariable competing-risk Cox regression model with 30-days readmission (requiring hospitalization) being the event of interest and deaths the competing event.
Results:Of 3,222 patients with Gram-negative bacteremia, 2,196 survived to discharge and 432 (19.6%) were readmitted within 30 days of discharge. Mean patients' age was 70+/-16 years. The bacteremia was hospital-acquired in 1,082 (49.3%) and healthcare-associated in 168 (7.7%). The most common pathogens isolated in blood were E. Coli, (997, 45.4% of patients). Cephalosporin or carbapenem resistance was observed among 119/338 (35.2%) of readmitted patients compared to 352/1,498 (23.5%) patients without readmission. The covariates retained in the final model predicting 30-day readmissions included nationality, active malignancy, pre-discharge heart rate, discharge with a nasogastric tube, immunosuppressive therapy, anticoagulation therapy, anxiolytic therapy, cephalosporin resistance, and total duration of antibiotic therapy. Appropriate empirical antibiotic treatment was not associated with survivors' readmissions (Table). We plan to present validation of the model in a different hospital.
Conclusions:Risk factors for hospital readmission in the month following sepsis include mainly baseline comorbidities and functional characteristics. However, ESBL phenotypes also contributed to the post-bacteremia morbidity. The risk factors identified can be used for assessing a patient’s readiness for discharge and to target post-discharge interventions for those discharged.
Keyword(s): Gran-negative bacteremia, Readmission, Multidrug resistance