Faculty Bibliography

Background. Due to altered pharmacokinetics/pharmacodynamics in critically ill patients, administration of beta-lactams as EI provides better target attainment in therapeutic drug monitoring studies. To optimize meropenem (MER) dosing in patients with severe sepsis or septic shock, our antimicrobial stewardship program implemented a MER EI protocol in an 18-bed medical intensive care unit (MICU) in March 2014. Methods. We conducted a retrospective evaluation to compare outcomes in MICU patients with severe sepsis and septic shock who received MER for >=72h administered as EI 1 g over 3 h Q8H with a total daily dose (TDD) 3g (1/2015-1/2017) vs. SI 500 mg over 30 minutes Q6H with TDD 2 g (1/2012-1/2014). ICU mortality and clinical response (CR) were evaluated as endpoints. CR was defined by improvement in signs and symptoms of infection. Results. Of 667 patients who received MER, 148 were included (EI n = 52, SI n = 96). Age, weight, comorbidities (malignancy 31 vs. 33%, P = 0.8; chronic liver disease [CLD] 15 vs. 23%, P = 0.4), severity of illness (median mAPACHE II 18 vs. 19, P = 0.6; SOFA 5 vs. 6, P = 0.5) and vasopressors' use (75 vs. 79%, P = 0.5) were comparable between EI and SI groups. Serum creatinine (SCr) was lower in EI group (median 1.1 vs. 1.4 SI, P = 0.05). Gram-negative (GN) pathogens (MIC<=0.25 mg/mL, 94%) were identified in 44% of patients in EI vs. 38% in SI group, P = 0.5. MER TDD was higher in EI group (3 vs. 1.5g SI, P < 0.01) with no difference in use of combination therapy (64 vs. 46%, P = 0.06). ICU mortality (median time to death 9 days) was lower (19 vs. 37%, P = 0.047) and CR was higher (83 vs. 46%, P = 0.038) in EI vs. SI group. Total pressor days on MER were shorter (2 vs. 3 days, P < 0.01) and white blood cell normalization rate was higher (87% vs. 51%, P < 0.01) in EI vs. SI group, whereas there was no difference in days of mechanical ventilation, duration of MER therapy and ICU stay. After adjusting for SCr, severity of illness scores, combination therapy and SI group in a multivariate model, CLD (OR 3.3, 95% CI 1.36-7.77, P = 0.008) and lower MER TDD (OR 1.8, 95% CI 1.09-2.98, P = 0.02) were independent predictors of ICU mortality. Conclusion. In this cohort of MICU patients with severe sepsis or septic shock and low MIC of GN pathogens, there was improved mortality and CR in MER EI group. Our finding of potential benefit of higher MER TDD in these patients warrants further exploration in a prospective study

Background. It is still unclear whether prolonged duration of therapy (DOT) for VAT might be protective against progression to pneumonia. From a stewardship view, shortening DOT may help to contain emergence of multidrug-resistant organisms (MDRO) in PICU. To this effect, we sought to compare clinical characteristics and outcomes in PICU patients with NLFGNR VAT treated with >7 days (prolonged course group, PCG) vs. <=7 days (short course group, SCG). Methods. This retrospective stewardship evaluation between January 2009 and July 2016 was conducted in a 12-bed PICU. Antibiotic choice and DOT were at the physicians' discretion. VAT was defined by signs and symptoms and positive sputum (>=moderate polymorphonuclear cells and >=moderate NLFGNR growth) without radiographic findings. Primary outcomes were rate of microbiologically documented or clinically suspected (CS) pulmonary infection recurrence and emergence of resistance (>=4 increase in minimal inhibitory concentration) or MDRO within 30 days of VAT treatment. Thirty-day readmission and in-hospital mortality were also assessed. Results. Fifty patients were included (PCG n = 27, SCG n = 23). Median age was 1.6 years (0-18.8), PIM2 score was 1 (0.1-82.8), 62% of patients had a tracheostomy at baseline, 70% had P. aeruginosa, and these were comparable between groups. More patients in PCG vs. SCG (44% vs. 13%, P = 0.03) had an admission diagnosis of respiratory failure. Mechanical ventilation (12.5 vs. 5 days, P < 0.01) and PICU stay (16 vs. 6 days, P < 0.01) were longer in PCG vs. SCG. Median DOT was 10 (8-30) in PCG vs. 6 days (3-7) in SCG, with beta-lactams as the common agents and no difference in combination therapy (33% vs. 13%, P = 0.1). Clinical response at the end of treatment was 89% in PCG and 100% in SCG, P = 0.2. Recurrence was 26% in PCG and 9% (all CS) in SCG, P = 0.2 at 17 days (1-29) and 9.5 days (4-15) P = 0.5, respectively. Emergence of resistance or MDRO occurred in 15% in PCG vs. 0% in SCG, P = 0.1. Readmission and in-hospital mortality were 7% vs. 9%, P = 0.9 and 7% vs. 0%, P = 0.5 in PCG and SCG, respectively. Conclusion. In this small cohort of PICU patients with NLFGNR VAT, there was no microbiologically documented recurrence and emergence of resistance or MDRO in SCG compared with PCG. Our findings suggest that short DOT may be considered for children who are less sick including those with a tracheostomy at baseline

PURPOSE: The integration of pharmacy residents into an antimicrobial stewardship program (ASP) is described, and data on the residents' ASP interventions and outcomes are reported. SUMMARY: ASP coverage of nighttime, holiday, and weekend shifts is often provided by infectious diseases (ID) medical fellows and staff pharmacists, potentially leading to inconsistent stewardship practices. As part of an initiative by a large urban hospital to provide around-the-clock, comprehensive ASP services 7 days a week, postgraduate year 2 (PGY2) pharmacy residents in ID or critical care were assigned to provide ASP coverage on weekends. Over a 12-month period, residents providing ASP weekend coverage documented a total of 1,443 interventions, of which 1,000 (69%) were pursuant to 72-hour prospective audit and feedback review and 443 (31%) occurred during ASP phone coverage. A comparison of overall antimicrobial utilization (mean +/- S.D. days of therapy [DOT] per 1,000 patient-days [PD]) before and after implementation of resident ASP coverage on weekends showed a decrease in aggregate antimicrobial use from 799.3 +/- 46.8 to 740.7 +/- 17.3 DOT/1,000 PD (a difference of 58.6 DOT/1,000 PD, p = 0.08), with a corresponding decline in the incidence of hospital-onset Clostridium difficile infection (from 1.18 cases to 0.9 case per 1,000 PD). CONCLUSION: By expanding the hospital's ASP services by assigning PGY2 pharmacy residents to weekend coverage, the institution was able to provide high-level clinical care 7 days per week, which benefited both patients and PGY2 pharmacy residents while meeting national ASP regulatory requirements.

BACKGROUND: Opioid-induced constipation (OIC) is common in critically ill patients; it leads to complications that can increase hospital stay and, rarely, bowel perforation. Opioid antagonists are considered a logical approach to treat OIC; however, the agent of choice has yet to be determined. OBJECTIVE: To assess the effectiveness and safety of enteral naloxone (NTX) versus subcutaneous methylnaltrexone (MNTX) for the treatment of OIC in the medical intensive care unit. METHODS: This retrospective review evaluated patients who received fentanyl continuous infusions for at least 72 hours and were initiated on NTX or MNTX. Active colitis, mechanical gastrointestinal obstruction, and inability to receive NTX orally were exclusion criteria. The primary outcome was time to first bowel movement (BM). Secondary outcomes included total number of BMs within 48 hours, opioid requirements after NTX or MNTX, and change in any of the following after the opioid antagonist: heart rates, mean arterial pressures, and level of sedation. A post hoc subgroup analysis of patients on vasopressors was conducted. RESULTS: Baseline characteristics were similar between patients receiving NTX (n = 52) and MNTX (n = 48), except the MNTX group required a higher median norepinephrine dose (0.22 vs 0.1 microg/kg/min, P = 0.001). The median times to first BM for NTX and MNTX were 30 and 24 hours ( P = 0.165). There was no difference in the primary outcomes for patients on vasopressors. Both groups did not require additional fentanyl equivalents, as evidenced by stable vitals and opioid requirements during treatment. CONCLUSIONS: Both agents appear to be effective and safe for the treatment of OIC; however, future controlled prospective trials are warranted.

STUDY OBJECTIVE: Despite recent reports of relatively high rates (16-37%) of acute kidney injury (AKI) in patients receiving the combination of intravenous piperacillin-tazobactam (PTZ) and vancomycin, data are limited evaluating the impact of PTZ infusion strategy on the occurrence of nephrotoxicity. The objective of this study was to compare the rates of nephrotoxicity in patients receiving vancomycin in combination with PTZ administered as an extended infusion (EI) versus a standard infusion (SI). DESIGN: Single-center, retrospective, matched-cohort study. SETTING: Large academic tertiary care hospital. PATIENTS: Two hundred eighty adults with a creatinine clearance (CrCl) of 40 ml/minute or higher who received at least 96 hours of vancomycin plus PTZ EI (140 patients) or vancomycin plus PTZ SI (140 patients) between January 1, 2009, and December 31, 2011, and between January 1, 2013, and December 31, 2014 (year 2012 was skipped due the closure of inpatient units following Superstorm Sandy); 48 patients in each group were admitted to the intensive care unit. MEASUREMENTS AND MAIN RESULTS: The median age of all patients was 67 (interquartile range [IQR] 54-77) years, and CrCl was 75 (IQR 55-107) ml/minute. Nephrotoxicity was assessed by the risk, injury, failure, loss, and end-stage kidney disease (RIFLE) and Acute Kidney Injury Network (AKIN) criteria. Rates of AKI, according to these criteria, were similar between groups: 17.9% versus 17.1% (p=1) and 32.9% versus 29.3% (p=0.596) for the PTZ EI and PTZ SI groups, respectively. When controlling for residual differences between groups in a conditional logistic regression analysis, no association was observed between receipt of PTZ EI and RIFLE-defined AKI (odds ratio 0.522, 95% confidence interval 0.043-6.295, p=0.609). Time to onset of nephrotoxicity was 4 (IQR 3-6) days, with no significant difference noted between groups (p=0.887). CONCLUSION: Our findings suggest a similar rate of nephrotoxicity between patients who received vancomycin in combination with PTZ EI versus PTZ SI. These results need to be further validated in a prospective randomized controlled study.

OBJECTIVE: Adrenergic crises are a cardinal feature of familial dysautonomia (FD). Traditionally, adrenergic crises have been treated with the sympatholytic agent clonidine or with benzodiazepines, which can cause excessive sedation and respiratory depression. Dexmedetomidine is a centrally-acting alpha 2-adrenergic agonist with greater selectivity and shorter half-life than clonidine. We evaluated the preliminary effectiveness and safety of intravenous dexmedetomidine in the treatment of refractory adrenergic crisis in patients with FD. METHODS: Retrospective chart review of patients with genetically confirmed FD who received intravenous dexmedetomidine for refractory adrenergic crises. The primary outcome was preliminary effectiveness of dexmedetomidine defined as change in blood pressure (BP) and heart rate (HR) 1 h after the initiation of dexmedetomidine. Secondary outcomes included incidence of adverse events related to dexmedetomidine, hospital and intensive care unit (ICU) length of stay, and hemodynamic parameters 12 h after dexmedetomidine cessation. RESULTS: Nine patients over 14 admissions were included in the final analysis. At 1 h after the initiation of dexmedetomidine, systolic BP decreased from 160 +/- 7 to 122 +/- 7 mmHg (p = 0.0005), diastolic BP decreased from 103 +/- 6 to 65 +/- 8 (p = 0.0003), and HR decreased from 112 +/- 4 to 100 +/- 5 bpm (p = 0.0047). The median total adverse events during dexmedetomidine infusion was 1 per admission. Median hospital length of stay was 9 days [interquartile range (IQR) 3-11 days] and median ICU length of stay was 7 days (IQR 3-11 days). CONCLUSIONS: Intravenous dexmedetomidine is safe in patients with FD and appears to be effective to treat refractory adrenergic crisis. Dexmedetomidine may be considered in FD patients who do not respond to conventional clonidine and benzodiazepine pharmacotherapy.

BACKGROUND: Cefepime and meropenem are used frequently in hospitalized patients for broad-spectrum empiric coverage, however, practitioners are often reluctant to prescribe these antibiotics for patients with a self-reported nonsevere, nontype I allergic reaction to penicillin. METHODS: Retrospective review of electronic medical records of adults with a self-reported allergy to penicillin who received at least 1 dose of cefepime, ceftriaxone, cefoxitin, cephalexin, or meropenem to assess incidence and type of allergic reactions. RESULTS: Of 175 patients included, 10 (6%) patients experienced an allergic reaction. The incidence for individual study drugs were cefepime 6% (6 of 96), meropenem 5% (3 of 56), cefoxitin 8% (1 of 13), ceftriaxone 0% (0 of 69), and cephalexin 0% (0 of 8). The majority of patients experienced a rash with or without pruritus and fever. Patients with a concomitant "sulfa" allergy (odds ratio [OR] 5.4, 95% confidence interval [CI] 1.4-21, P = .02) or >/=3 other drug allergies (OR 6.4, 95% CI 1.3-32, P = .025) were more likely to have an allergic reaction. CONCLUSIONS: In one of the largest retrospective reviews of hospitalized patients who received full dose therapy with cefepime, ceftriaxone, and meropenem, the incidence of allergic reactions was low and reactions were mild. Cefepime, ceftriaxone, and meropenem can be considered for use in patients with a self-reported nontype I penicillin allergy.

Vasopressors are an integral component of the management of septic shock and are traditionally given via a central venous catheter (CVC) due to the risk of tissue injury and necrosis if extravasated. However, the need for a CVC for the management of septic shock has been questioned, and the risk of extravasation and incidence of severe injury when vasopressors are given via a peripheral venous line (PVL) remains poorly defined. We performed a retrospective chart review of 202 patients who received vasopressors through a PVL. The objective was to describe the vasopressors administered peripherally, PVL size and location, the incidence of extravasation events, and the management of extravasation events. The primary vasopressors used were norepinephrine and phenylephrine. The most common PVL sites used were the forearm and antecubital fossa. The incidence of extravasation was 4%. All of the events were managed conservatively; none required an antidote or surgical management. Vasopressors were restarted at another peripheral site in 88% of the events. The incidence of extravasation was similar to prior studies. The use of a PVL for administration of vasopressors can be considered in patients with a contraindication to a CVC.