Enhanced antimicrobial de-escalation for pneumonia in mechanically ventilated patients: a cross-over study

The study was conducted within the two medical ICUs (total 34 beds). at Barnes-Jewish Hospital, an academic referral center of 1250 beds. The medical ICUs are geographically co-located closed units with shared physician, nursing, pharmacist, and respiratory therapist staff. The medical ICUs are staffed 24 hours per day and 7 days per week by these teams, including intensivists board-certified in critical care. This investigation was approved by the Washington University School of Medicine Human Studies Committee and the need for informed consent was waived (Institutional Review Board (IRB)# 201509075; ClinicalTrials.gov Identifier: NCT02685930). All mechanically ventilated patients with suspected pneumonia, from 1 January 2016 through 31 December 2016 were eligible for inclusion. Patients were excluded if they had a concomitant non-pulmonary infection requiring antibiotic therapy, bronchiectasis, or significant immune suppression. Data were prospectively collected from the electronic health record and from patients’ ICU teams.

The primary objective of this study was to determine whether and how frequently antibiotic de-escalation occurred following the initial prescription of antibiotics for suspected pneumonia. The secondary outcomes assessed included antibiotic duration, failure of initial antibiotic therapy, hospital mortality, ICU and hospital duration, and subsequent infection with antibiotic-resistant strains of bacteria.

The two medical ICUs were initially randomly assigned so that subjects admitted to each ICU would receive either routine antibiotic management (RAM) for suspected pneumonia or EAD. Patients in the RAM group had all antibiotic decisions determined by their ICU team composed of a critical care board-certified attending physician, a critical care fellow, resident physicians, and a critical care specialty pharmacist. Antibiotic decision-making was similar in patients in the EAD group except that all antibiotic orders and microbiology results were reviewed on a daily basis by the study team (TT, KF, STM, PJ, and MHK) to specifically determine if further antibiotic de-escalation could occur. The study team performed these reviews from 7 a.m. to 6 p.m. on weekdays excluding weekends and holidays. After 6 months the two medical ICUs were crossed over in terms of the type of antibiotic management strategy delivered to patients.

Adult patients (age >18 years) were identified prospectively with suspected pneumonia (community-onset pneumonia (COP), hospital-acquired pneumonia (HAP), or ventilator-associated pneumonia (VAP)) in accordance with the American Thoracic Society position statement on pneumonia [11]. The diagnosis included presence of a new or progressive radiographic infiltrate and at least two of the following clinical features: fever >38 °C, leukocytosis (>10 × 10⁹cells/L), leukopenia (≤4 × 10⁹cells/L), or purulent secretions. The presence of a new or progressive radiographic infiltrate was based on the interpretation of the chest radiograph by board-certified radiologists blinded to the study. Patients admitted from the community without prior radiographs were assumed to have a new infiltrate when present. Pneumonia was classified microbiologically according to the pathogen(s) identified as pathogen-negative, viral, antibiotic-susceptible, or antibiotic-resistant. For purposes of this investigation, antibiotic-susceptible was determined according to ceftriaxone susceptibility, as ceftriaxone represents the antimicrobial agent most frequently recommended for hospitalized patients with pneumonia coming from the community setting [12]. Additionally, patients could be classified as having non-infectious illness mimicking suspected pneumonia (see Additional file 1: Table S1). Patients classified as having viral pneumonia had to meet the aforementioned clinical and radiographic criteria for pneumonia.

Early failure of initial antibiotic therapy was defined as progression to shock, mortality, or progression of the radiographic infiltrates without resolution of clinical parameters (leukocytosis or leukopenia, fever, purulent secretions) within 48 hours of starting antibiotic therapy. Septic shock was defined as the need for vasopressors (norepinephrine, dopamine, vasopressin, epinephrine, or phenylephrine). Only the first episode of suspected pneumonia was recorded. Antimicrobial treatment was classified as IAAT if the initial regimen had in vitro activity demonstrated against the isolated pathogens. Immune suppression was defined as acquired immune deficiency syndrome, solid organ or bone marrow transplant, hematologic malignancies, solid cell cancers treated with chemotherapy or radiation, long-term corticosteroids (>10 mg/day), and other immunosuppressive drugs (e.g., biologics for rheumatologic disorders). Multidrug-resistant (MDR) pathogens had to demonstrate in vitro resistance to at least one agent from three distinct classes of antimicrobials that would normally have activity against that bacterium [13].

Antimicrobial de-escalation was defined as a change in the initially prescribed antibiotic regimen resulting in one of the following: (1) a reduction in the number of antibiotics prescribed, (2) elimination of antibiotic coverage for a class of pathogens (e.g., stopping coverage for Staphylococcus aureus), and (3) changing the antibiotic regimen to a more narrow-spectrum regimen. For purposes of this investigation, antibiotics were ranked according to their activity spectrum against pneumonia pathogens, emphasizing Gram-negative bacteria (5, highest; 0, lowest) and Gram-positive bacteria (1, highest; 0, lowest) (Table 1). For combination regimens, rank was assigned according to the most potent drug. Therapy escalation was defined as the switch to or addition of a drug class or classes with a broader spectrum (using definitions in Table 1).

From January 2002 through the present, Barnes-Jewish Hospital utilized an antibiotic control program to help guide antimicrobial therapy for bacterial infections. During this time, the use of cefepime, gentamicin, or vancomycin was unrestricted. However, initiation of intravenous ciprofloxacin, imipenem, meropenem, piperacillin/tazobactam, ceftolozone/tazobactam, ceftazidime/avibactam, linezolid, ceftaroline, or daptomycin was restricted and required preauthorization from either a clinical pharmacist or infectious diseases physician. Each intensive care unit (ICU) had a clinical pharmacist who reviewed all antibiotic orders to ensure that dosing and interval of antibiotic administration was adequate for individual patients based on body size, renal function, and the resuscitation status of the patient. The duration of antibiotic therapy was at the discretion of the ICU treating team based on the patient’s microbiology results and response to therapy.

The initial antibiotic dosages employed for the treatment of bacterial infections at Barnes-Jewish Hospital were as follows: cefepime, 1 to 2 g every 8 hours; piperacillin–tazobactam, 4.5 g every 6 hours; imipenem, 0.5 g every 6 hours; meropenem, 1 to 2 g every 8 hours; ceftolozone/tazobactam, 1.5 or 3 g every 8 hours; ceftazidime/avibactam, 2.5 g every 8 hours; ciprofloxacin, 400 mg every 8 hours; levofloxacin, 750 mg once daily; vancomycin, 15 mg/kg every 12 hours; linezolid, 600 mg every 12 hours; and ceftaroline, 600 mg every 8 hours. Subsequent antibiotic dose and frequency were adjusted, where appropriate.

The microbiology laboratory performed antimicrobial susceptibility of the bacterial isolates using the disk diffusion method according to guidelines and breakpoints established by the Clinical Laboratory and Standards Institute and published during the inclusive years of the study [14]. All classifications of antibiotic resistance were based on in vitro susceptibility testing using these established breakpoints. Viral identification was obtained using the Biofire FilmArray® Respiratory Panel (bioMérieux, Marcy l'Etoile, France).

The sample size was determined by a convenience sample of all mechanically ventilated patients with suspected pneumonia identified in the medical ICUs during the study period. Continuous variables were expressed as mean and standard deviation (SD) or median and interquartile range (IQR) when appropriate. The t test was used to analyze normally distributed continuous variables, whereas the Mann–Whitney U test was used to analyze non-normally distributed continuous variables. Categorical data were reported as frequency distributions and analyzed using the chi-square test. P values less than 0.05 were considered statistically significant, and all tests were two-tailed. All analyses were done using SPSS Statistics 21 (IBM SPSS Statistics, Version 21.0. Armonk, NY, USA).