Impact of Factor Xa Inhibitor Reversal with Prothrombin Complex Concentrate in Patients with Traumatic Brain Injuries.

BACKGROUND: Anticoagulant use prior to trauma has been associated with increased incidence of traumatic brain injury (TBI), intracranial hemorrhage (ICH) progression, and mortality. Prothrombin complex concentrates (PCCs) are commonly used as off-label treatments for factor Xa inhibitor-associated life-threatening hemorrhage. At this time, there is no consensus regarding appropriate indication, target dose, or outcomes of PCC administration in patients presenting with traumatic ICH. This study seeks to evaluate the impact of reversal with PCC on hemorrhage progression and outcomes in patients with TBI on preinjury factor Xa inhibitors. METHODS: This single-center retrospective cohort study included patients ≥ 18 years presenting with an acute TBI of any severity on apixaban or rivaroxaban from September 1, 2016, to September 1, 2019. Patients were grouped on the basis of receipt of PCCs for reversal (i.e., reversal or no reversal). Exclusion criteria included spontaneous ICH or known coagulopathy. Propensity score matching was conducted with the following variables: age, Abbreviated Injury Scale (head) score, and Charlson Comorbidity Index score. The primary outcome was hemorrhage stability within 48 h. Secondary outcomes included degree of hemorrhage progression, in-hospital mortality, discharge disposition, and incidence of thromboembolic events. RESULTS: Of the 115 patients meeting inclusion criteria, 84 were included in the propensity score matched data set. Baseline characteristics, comorbidities, and TBI severity were similar. The majority of patients in the reversal group (35 [83.3%]) and the no reversal (NR) group (40 [95.2%]) experienced a mild TBI (admission Glasgow Coma Scale score of 14 to 15). In the reversal group, patients received 34.3 units/kg activated PCC, 30.5 units/kg four-factor PCC, or 54.9 units/kg four-factor PCC and activated PCC on average. There was no difference observed in the incidence of hemorrhage progression (10.8% NR vs. 15.0% reversal; p = 0.739) or in median change in ICH volume (0 mL NR vs. 1 mL reversal; p = 0.2199) between groups. Additionally, reversal did not affect in-hospital mortality (3 [7.1%] NR vs. 4 [9.5%] reversal; p > 0.999). One patient in the reversal group developed a deep vein thrombosis (DVT) during the hospitalization; however, this did not result in a statistically significant difference in the occurrence of DVT (p > 0.999). CONCLUSIONS: This study demonstrated that PCC used for the treatment of factor Xa inhibitor-associated ICH related to mild TBI did not significantly impact the incidence or degree of hemorrhage progression, and PCC treatment did not result in increased thromboembolic events.

Impact of a Plan of Care Protocol on Patient Outcomes in People Who Inject Drugs With Infective Endocarditis.

BACKGROUND: People who inject drugs (PWID) are at increased risk of deleterious sequelae due to infective endocarditis (IE). A standardized, hospital-wide drug use-associated infection protocol targeting medication safety, pain management, and limiting external risk factors was implemented at an academic medical center to improve outcomes in PWID with IE. METHODS: A quasi-experimental study included patients with active injection drug use and definite IE from January 2013 to July 2017 (preintervention group) and from September 2017 to January 2019 (intervention group). The primary outcome of interest was the 90-day all-cause readmission rate. Secondary outcomes included infection-related readmission rates, in-hospital and all-cause mortality rates, and the frequency of patients leaving against medical advice. RESULTS: A total of 168 patients were included, in the 100 preintervention and 68 in the intervention group. Patients in the intervention group had reduced odds of 90-day all-cause readmission (adjusted odds ratio, 0.2; 95% confidence interval, 0.08-0.6) after adjustment for confounding variables. The 12-month all-cause mortality rate was also significantly reduced in the intervention group (adjusted odds ratio, 0.25; 95% confidence interval, .07-.89). The intervention group had a higher proportion of patients leaving against medical advice (6% for the preintervention group vs 35% for the intervention group, P < .001). CONCLUSIONS: A drug use-associated infection protocol demonstrated reduced 90-day all-cause readmission and 12-month all-cause mortality rates in PWID with IE. This study highlights the importance of standardized care processes for improving care in this specialized patient population.

Epidemiology and outcomes of non-HACEK infective endocarditis in the southeast United States.

OBJECTIVES: Infective endocarditis (IE) with non-HACEK Gram-negative (GN) organisms is rare, but associated with poor outcomes. The purpose of this study was to quantify the microbiology, treatment strategies, and frequency of poor outcomes in patients with non-HACEK GN IE. MATERIALS: Retrospective cohort of adults with definite non-HACEK GN IE from 1/11-1/19. The primary endpoint was poor patient outcome, defined as a composite of all-cause death or infection-related readmission within 90-days of index infection. RESULTS: 43 patients were included: 51% patients were men, and the median (IQR) age was 40 (31-50) years. Forty patients reported injection drug use. The most common organisms were Pseudomonas aeruginosa (68%) and Serratia marcescens (9%). Seventy-six percent of patients received definitive combination therapy; the most common antibiotics used in combination with a ß-lactam were aminoglycosides (50%) and fluoroquinolones (34%). Three patients discontinued combination therapy due to toxicity. Twelve-month, all-cause mortality and readmission was 30% and 54%, respectively. In multivariable logistic regression, variables independently associated with composite poor outcome were receipt of fluoroquinolone-based IE combination therapy and septic shock. CONCLUSIONS: Long-term mortality and readmission rates were high. Patients who received fluoroquinolone-based IE combination therapy more frequently developed poor outcomes than those who did not.