Faculty Bibliography

WHAT IS KNOWN AND OBJECTIVE/OBJECTIVE:There is limited guidance on how to transition critically ill patients from intravenous (IV) to oral (PO) amiodarone. The objective of this study was to assess the impact of IV and PO amiodarone overlap on short-term tachyarrhythmia recurrence and adverse hemodynamic outcomes in the intensive care unit. METHODS:This is a retrospective, single-center analysis of critically ill adults who were treated with IV amiodarone for a supraventricular arrhythmia with rapid ventricular rate (RVR) and transitioned to PO amiodarone while inpatient. Patients were excluded if rate control was not achieved prior to the PO transition. Receipt of concomitant IV and PO therapy for ≤2 hours was considered no overlap (NOV) and >2 hours was considered overlap (OV). Tachyarrhythmia recurrence and adverse hemodynamic events were compared between groups. RESULTS:A total of 90 patients (45 NOV, 45 OV) were included in the analysis. The median overlap duration was 0.1 (-1.3 to 1.2) hours in the NOV arm and 4 (2.6-6.1) hours in the OV arm. Recurrence of RVR occurred in 9 (20%) patients in each arm (P = 1.0). The median time from IV discontinuation to return of tachyarrhythmia was 10.5 hours. There were no significant differences in amiodarone dosing, electrolyte abnormalities, volume status or concomitant cardiac medications at the time of IV to PO transition. Hypotension occurred in 13% and 20% (P = 0.369) and bradycardia in 9% and 13% (P = 0.502) of patients in the NOV and OV arms, respectively. WHAT IS NEW AND CONCLUSION/CONCLUSIONS:Providing IV and PO overlap of amiodarone for a median of 4 hours did not decrease the rate of early tachyarrhythmia recurrence. Future studies are warranted to evaluate the impact of alternative amiodarone dosing strategies on breakthrough tachyarrhythmia.

Background: Midline catheters (MCs) have arisen as alternatives to peripherally inserted central catheters (PICCs) for both general intravenous therapy and extended outpatient parenteral therapy. However, there is a lack of data concerning the safety of medication therapy through midline for extended durations. Objective: The purpose of this study is to evaluate the safety of MCs for extended intravenous use. Methods: This was a retrospective cohort study evaluating patients who received intravenous therapy through an MC at a tertiary care academic medical center. The primary end point was the incidence of composite catheter-related adverse events that included local events, catheter dislodgment, infiltration, catheter occlusion, catheter-related venous thromboembolism, extravasation, and line-associated infection. Results: A total of 82 MC placements and 50 PICC placements were included; 50 MCs were for outpatient parenteral antimicrobial therapy, and 32 were for inpatient intravenous use. There were 21 complications per 1000 catheter-days in the outpatient group and 7 complications per 1000 catheter-days in the PICC group (P = 0.91). The median time to complication in both groups was 8 days. The antimicrobial classes commonly associated with complications were cephalosporins, carbapenems, and penicillins. Conclusion and Relevance: Our results suggest that intravenous therapy with MCs is generally safe for prolonged courses that do not exceed 14 days as compared with PICC lines, which can be placed for months. There is still limited evidence for the use of MCs between 14 and 28 days of therapy. This study can help guide our selection of intravenous catheters for the purpose of outpatient antimicrobial therapy.

Background:There is a paucity of data evaluating optimal dosing strategies of commonly utilized opioids and sedatives for patients on extracorporeal membrane oxygenation (ECMO) support where pharmacokinetic and pharmacodynamic variables of these administered agents are altered. Objective: To assess the daily dosing requirement of sedatives and analgesics for patients on venovenous (VV) and venoarterial (VA) ECMO after the initial ECMO cannulation period. Methods: We performed a retrospective, observational study of adult patients receiving sedation and analgesia while receiving ECMO support for at least 24 hours. Patients cannulated at an outside hospital more than 24 hours before transfer, those with a history of intravenous drug use or acute alcohol withdrawal, or those who died within 48 hours of ECMO initiation were excluded. Results: We evaluated 26 patients on ECMO, including 13 on VV and 13 on VA ECMO. The median dose of fentanyl was 140 µg/h, with the VV group requiring a higher dose compared with the VA group (167 vs 106 µg/h, P < 0.001). The median doses of dexmedetomidine and propofol were 0.7 µg/kg/h and 26 µg/kg/min, respectively, with no significant differences between groups (P = 0.38 and P = 0.24, respectively). The median daily doses of fentanyl, dexmedetomidine, and propofol did not significantly increase throughout the time on ECMO support. Conclusions and Relevance: We found that the overall opioid daily dosing requirements were lower than previously reported in the literature. Additionally, light sedation strategies with a target RASS of -1 to 0 are feasible in this patient population.

BACKGROUND:Although dalbavancin's (DBV's) long half-life and one-time dosing strategy confer ideal administration in the ambulatory setting, the optimal role of DBV in the management of acute bacterial skin and skin structure infections (ABSSSIs) remains to be elucidated. OBJECTIVES/OBJECTIVE:The primary objective of this study was to compare treatment outcomes of ABSSSI between patients who received DBV in the emergency department (ED) as part of standard care versus patients who received DBV as part of a telehealth program. METHODS:This was a retrospective cohort study evaluating patients who received DBV at 3 urban EDs. The primary end point was the incidence of ABSSSI recurrence. Secondary outcomes included need for hospital admission and ED length of stay (LOS; in hours). RESULTS:A total of 65 ABSSSI treatment courses were included; 42 were included in the telehealth criteria (TC) cohort and 23 in the initial criteria (IC) cohort. There were 14% (6/42) infection recurrences in the TC cohort and 22% (5/23) in the IC cohort, with median time to recurrence being 4 and 14 days, respectively. Median ED LOS was significantly shorter in the TC (5 vs 25 hours, P < 0.05). Numerically fewer individuals in the TC cohort required inpatient admission (0 vs n = 2, 9%). Conclusion and Relevance: Our results suggest that patients may be safely administered DBV in an ED setting, with telehealth follow-up. Providing structured patient selection criteria is an effective method of assisting ED providers in selecting appropriate DBV candidates to limit potential recurrences and readmissions.

Objective Patients with acute myelogenous leukemia or myelodysplastic syndrome undergoing induction chemotherapy are at increased risk of invasive fungal infection due to prolonged, severe neutropenia. Due to this risk, national guidelines recommend invasive fungal infection prophylaxis in this population until the resolution of neutropenia. Although posaconazole has demonstrated superiority over fluconazole and itraconazole, there is limited evidence for voriconazole for invasive fungal infection prophylaxis in this population. Even less data are available comparing posaconazole and voriconazole directly. The study objective was to investigate the efficacy and safety of delayed-release posaconazole tablets versus voriconazole for primary invasive fungal infection prophylaxis. The primary outcome was rate of discontinuation of either agent. Secondary outcomes included specific rates of discontinuation due to adverse events and drug-drug interactions, rates of breakthrough invasive fungal infection, and 30-day and 100-day mortality rates. Methods This was a retrospective cohort study of adult patients admitted to NYU Langone Health between 1 January 2014 and 31 August 2017 and initiated on invasive fungal infection prophylaxis during induction or reinduction chemotherapy for acute myelogenous leukemia or myelodysplastic syndrome. Results In total, 77 patients were included in the study: 43 using posaconazole delayed-release tablets and 34 using oral voriconazole. In the posaconazole group, 30% of patients (n = 13) discontinued therapy for any reason compared with 35% (n = 12) of patients in the voriconazole group (p = 0.64). A higher percentage of patients in the voriconazole group discontinued due to adverse events (6 patients, 18% vs. 1 patient, 2%, p = 0.04). Mortality rates at 30 and 100 days were similar between both groups. No breakthrough invasive fungal infections was noted in either group. Conclusion Overall, discontinuations for any reason were similar in patients taking both posaconazole delayed-release and oral voriconazole. Both posaconazole delayed-release tablets and oral voriconazole appear to be effective at preventing invasive fungal infection in acute myelogenous leukemia and myelodysplastic syndrome patients undergoing induction chemotherapy, although posaconazole may be more tolerable.

Carfilzomib is a second-generation proteasome inhibitor that irreversibly inhibits chymotrypsin-like (CT-L) activities of the proteasome, and is indicated for relapsed or refractory multiple myeloma. Cardiotoxicity is a well-established adverse effect of carfilzomib. The extent of cardiac toxicity in the literature spans anywhere from palpitations to cardiac arrest, with the most commonly reported manifestation being new-onset or worsening heart failure. A pre-clinical study of the pharmacokinetics and pharmacodynamics of carfilzomib given via intravenous bolus or 30-minute infusion in rats showed that carfilzomib can strongly induce apoptosis and potently damage cardiac myocytes at clinically relevant concentrations. Moreover, the mortality rate with the bolus administration was 44% whereas the same dose administered as a 30-minute infusion did not result in mortality. There remains limited clinical data regarding the safety of carfilzomib at doses of 27-56 mg/m2 based on infusion times as these doses have not been well studied. This retrospective review was conducted to evaluate the safety of carfilzomib at doses >27 mg/m2 at all infusion times.

Limited evidence is available to guide periprocedural management of oral anticoagulants in the setting of interventional radiology (IR) procedures. For direct oral anticoagulants, therapy interruption (TI) is based on medication half-life and procedural bleeding risk. Periprocedural management of warfarin includes INR monitoring, and possible bridging with parenteral anticoagulants. It is unknown if these recommendations apply to IR procedures. To evaluate bleeding complications and thromboembolic events following periprocedural management of the factor Xa (FXa) inhibitors or warfarin in patients undergoing IR procedures. We performed a retrospective, observational study at NYU Langone Health (NYULH) of all adult patients who underwent an IR procedure from January 2015 to July 2017 and were receiving apixaban, rivaroxaban, or warfarin. Patients who were pregnant or who had a mechanical heart valve were excluded. At NYULH, TI is not required for FXa inhibitors, and an INR < 3 is recommended for patients on warfarin undergoing low risk procedures. For moderate/high risk procedures, TI for 48 h or 72 h with reduced renal function, is recommended for FXa inhibitors, and an INR < 1.5 is recommended for patients on warfarin. We evaluated 350 IR procedures, with a total of 174 low bleeding risk and 176 moderate/high bleeding risk. The 30-day major bleeding rate was 0.9%, clinically relevant non-major bleeding rate was 3%, minor bleeding rate was 1% and thromboembolic event rate was 1%. The periprocedural oral anticoagulation management strategy at NYULH appears safe given the low 30-day incidence of bleeding and thromboembolic events.

Background. Oral treatment strategies for Enterobacteriaceae bacteremia (EB) are controversial, with both beta-lactams (BL) and fluoroquinolones (FQ) used in clinical practice. FQ may be preferred for their high bioavailability, but other oral antibiotics are needed due to concerns of resistance and adverse effects. As an effort to facilitate antibiotic stewardship, BL should be explored as an additional oral option for EB treatment. Methods. This retrospective study compared clinical characteristics and outcomes in patients with EB treated with BL vs. FQ as definitive oral therapy between January 2013 and July 2017. Adult patients diagnosed with their first incidence of EB and transitioned from IV antibiotics to either study antibiotic class were included. Primary and secondary outcomes assessed recurrence, collateral damage, readmission, and all-cause mortality. Results. A total of 173 patients were included (BL n = 59, FQ n = 114). Median age was 70 years, Pitt bacteremia score was 2 (range 0-7), and Charlson Comorbidity Index was 5 (0-12); all were comparable between groups. Urinary source of infection was most common (57%). The majority of oral BL courses used cefpodoxime (63%). More patients in FQ vs. BL had a prior transplant (9% vs. 0%, P = 0.05), presence of abscess (11% vs. 0%, P = 0.01), and Infectious Diseases consultation (63% vs. 34%, P = 0.0001). Onset of EB in an intensive care unit was more common in BL vs. FQ (24% vs. 10%, P = 0.01). Median duration of IV and oral therapy was 5 vs. 4 days, P = 0.22 and 11 vs. 12 days, P = 0.17 in BL and FQ, respectively. Recurrence within 90 days was 7% in BL and 4% in FQ, P = 0.49 (adjusted OR 1.44, 95% CI 0.31-6.66; P = 0.64). Multivariate analysis identified liver cirrhosis (OR 16.89, 95% CI 1.06-268.32; P = 0.05) as an independent predictor of recurrence within 90 days. All secondary outcomes were similar between BL vs. FQ: superinfection within 90 days (10% vs. 9%, P = 0.76), C. difficile infection within 90 days (3% vs. 1%, P = 0.27), 30-day readmission (15% vs. 20%, P = 0.43), all-cause 30-day mortality (0% vs. 3%, P = 0.55). Conclusion. In our cohort of patients with EB, clinical outcomes were similar between those treated with oral BL compared with FQ. Oral BL may be considered for definitive treatment of EB, although further investigation in larger studies is needed