Background: Phenobarbital offers several possible advantages to benzodiazepines including a longer half-life and anti-glutamate activity, and is an alternative for the treatment of alcohol withdrawal. The objective of this analysis was to evaluate the safety and efficacy of a phenobarbital protocol for alcohol withdrawal newly implemented at our institution. Methods: This was a single-center, retrospective analysis of adult patients admitted to the medical/surgical/burn/trauma intensive care unit (ICU) with or at risk of severe alcohol withdrawal. Patients who were admitted prior to guideline implementation and received scheduled benzodiazepines (PRE) were compared to those who received phenobarbital post guideline update (POST). The primary outcome was ICU length of stay (LOS). Results: Upon analysis, 68 patients in the PRE and 64 patients in the POST were identified for inclusion. The median APACHE II score was significantly higher in the POST (4.5 [3:9] vs 10 [5:13], P < 0.001). ICU (2 [1:2] vs 2 [2:5], P = 0.002) and hospital (4.5 [3:6] vs 8 [6:12], P < 0.001) LOS were significantly longer in the POST. There was no difference in mortality or duration of mechanical ventilation. More patients required propofol or dexmedetomidine on day one in the POST ( P < 0.001). Conclusion: Patients in the POST had significantly longer ICU and hospital LOS, and had a higher baseline severity of illness. Future research is needed to evaluate the efficacy and safety of phenobarbital compared to benzodiazepines for severe alcohol withdrawal.

Aim: To discuss two patient cases of Serratia marcescens endocarditis and the paucity of literature regarding treatment options. Presentation of Case: Patient 1 was a 29-year old male who presented with native mitral valve Serratia marcescens endocarditis presumed secondary to intravenous drug use. He was empirically treated with vancomycin and piperacillin/tazobactam then transitioned to meropenem and gentamicin 1 mg/kg every 8 hours. He was maintained on vancomycin monotherapy for days 4-14. Gentamicin was restarted on hospital day 14 at 7 mg/kg every 36 hours for 6 weeks. He underwent mitral valve replacement on hospital day 20. He was readmitted on day 42 with splenic lesions and enlarging mycotic aneurysms. Patient 2 was a 38-year old male with native aortic valve Serratia marcescens endocarditis with septic emboli presumed secondary to intravenous drug use. He was treated with vancomycin and cefepime then was transitioned to ceftriaxone and levofloxacin. The patient underwent aortic valve replacement on hospital day 3 and was transitioned to meropenem and levofloxacin for 6 weeks. Discussion: The treatment strategies for both patients demonstrates that the optimal treatment strategy for Serratia marcescens endocarditis remains unclear. The gentamicin dosing for patient 1 demonstrates “synergy” and extended-interval dosing. Despite both dosing strategies being used, the patient continued to exhibit complications of the infection. Patient 2 demonstrates successful treatment of the infection with surgical intervention and a carbapenem/fluoroquinolone regimen. Conclusion: These cases demonstrates that much remains unclear in the treatment of Serratia marcescens endocarditis and more studies and case reports are needed.

Background: Antifibrinolytic agents, tranexamic acid (TXA) and epsilon-aminocaproic acid (EACA), are often used during cardiac surgery to decrease the number of allogenic blood transfusions and to prevent perioperative bleeding. Weight-based TXA dosing regimens have been compared to fixed-dose regimens of EACA with variable outcomes in perioperative blood product transfusions and chest tube output. Serious adverse events, including seizures, have been reported with higher doses of TXA. Fixed-dose TXA regimens have been evaluated in trauma and orthopedic surgery but there is a paucity of evidence in the cardiac surgery population. Aims of the Study:To compare the safety and efficacy of fixed-dose TXA versus EACA in patients undergoing cardiac surgery.Methods: A single-center, retrospective chart review was conducted at a 793-bed tertiary care academic teaching hospital comparing cardiac surgery patients receiving either fixed-dose TXA 1000 mg followed by a 500-1000 mg infusion or EACA-7.5 g intravenous boluses followed by a 1-1.25 g/h infusion for the duration of the surgery. The major endpoint included chest tube output at 12 h, 24 h, and 7 days postoperatively. Minor endpoints included quantity and incidence of blood product transfusions and reported safety events.Results: There were 1544 patients included. Chest tube output was similar between groups and the TXA group required more intraoperative blood product transfusions (22.7% vs. 18.2%, p = .03). There were no differences in the median quantity of total blood products administered postoperatively at 24 h or at 7 days. Reported safety events were similar between groups. Conclusion:Both fixed-dose TXA and EACA may be considered safe and effective options for antifibrinolytic therapy in cardiac surgery patients.

Background: Opioid-induced constipation (OIC) may occur in up to 81% of critically ill patients and can lead to many complications. Opioid antagonists are a reasonable approach and may be used for managing OIC. Objective: The purpose of this study was to assess the efficacy of enteral naloxone (NLX) versus subcutaneous methylnaltrexone (MNTX) for the management of OIC in critically ill patients. Methods: A retrospective analysis was conducted on adult patients who received NLX or MNTX and a continuous opioid infusion for at least 48 hours. The primary end point was time to resolution of constipation, defined as hours to first bowel movement (BM) after the first dose of an opioid antagonist. Reversal of analgesia was assessed by comparing the total number of morphine milligram equivalents (MME) 24 hours preopioid and postopioid antagonist administration. Univariate and multivariate analyses were conducted to assess treatment response within 48 hours. Results: Baseline characteristics were similar between patients receiving NTX (n = 89) and MNTX (n = 71). However, the time to the first BM with NLX was 18 hours compared with 41 hours with MNTX ( P = 0.004). There was no difference in MME requirements 24 hours pre/post NLX or MNTX administration. Naloxone administration was identified as a statistically significant predictor of BM within 48 hours (odds ratio [OR] = 2.68 [1.33-5.38]). Conclusion and Relevance: The time to first BM was shorter with enteral NLX. Both NLX and MNTX appear to be effective for the management of OIC without causing reversal of analgesia. Future controlled, prospective trials comparing these agents are warranted.

ObjectiveIn 2017, interns were permitted to work continuously for up to 28 hours at a time, a reversal from the previously mandated 16-hour limit. Our objective was to evaluate perceptions of care and patient outcomes on an extended (28-hour) compared with a limited (16-hour) duty-hour system on identical interdisciplinary teams.MethodsSixty-two interns, 27 residents, 28 attendings, and 449 patients participated. Patients completed surveys assessing their satisfaction. Anonymous weekly surveys were obtained from interns, residents, and attendings evaluating perceptions of intern tiredness, overall satisfaction, and performance. Nursing surveys evaluated intern and medical team performance. Objective outcome measures, including intensive care unit transfers, length of stay, readmissions, mortality, and complications, were assessed through a retrospective, blinded chart review.ResultsPatients reported similar satisfaction in care. Extended duty-hour interns reported significantly decreased familiarity with their patients, decreased ability to conduct physical exams on new patients, increased tiredness, and decreased overall satisfaction. Residents overseeing extended-duty interns reported significantly decreased quality in intern presentations and overall quality of teaching, and increased perception of intern tiredness and increased incorrect orders. Attending physicians reported significantly improved quality of new patient presentations by extended duty-hour interns. No significant differences in patient objective outcome measures were noted.ConclusionsExtended intern duty hours do not affect patient’s satisfaction with their care. Although interns in the extended duty-hour system reported significantly increased fatigue and decreased overall satisfaction and residents’ perceived increases in incorrect intern orders in the extended duty-hour system, there were no detrimental effects on patient safety.