1429. QI: Modified Clinical Pulmonary Infection Score as a surrogate tool to guide antibiotic de-escalation in patients with hospital-acquired, healthcare-associated and ventilator-associated pneumonia in intensive care units
Session: Poster Abstract Session: Antimicrobial Stewardship: Interventions
Saturday, October 10, 2015
Room: Poster Hall
  • Poster for ATS final.pdf (478.7 kB)
  • Background:

    The Clinical Pulmonary Infection Score (CPIS), was initially developed to assist in bedside diagnosis of ventilator associated pneumonia (VAP).1,2  Patients with a low clinical suspicion and CPIS <6 can have antibiotics safely discontinued after 3 days.3The disadvantages, however, are its high inter-observer variability and the need for arterial blood gas analysis for calculation.  As a result, the score cannot be used for a substantial proportion of pneumonia patients for whom ABGs are not available when de-escalation might be considered.


    Modified CPIS (MCPIS) was created.  2 components were revised: oxygenation (PaO2/FiO2) and sputum description. 200 patients admitted to the Yale-New Haven Hospital ICU with HCAP, HAP and VAP from 07/01/2011 to 06/31/2012 were randomly chosen. In Phase I, 2 investigators independently performed a retrospective chart review. The Original and MCPIS were calculated on admission and at 72 hours. In Phase II, an antimicrobial stewardship program was established with implementation of MCPIS to assist clinicians with antibiotic de-escalation.  Sputum collection and antibiotic de-escalation practices were assessed in both phases.


    85 patients were included in the study. Because of absent ABG results, we were unable to calculate 41% and 76% of the initial and 72 hour Original CPIS respectively. Of the patients with low Original CPIS score (<6), 43% received broad spectrum beyond 72 hours.  With the MCPIS we were able to calculate the score in 100% of the patients.  At 72 hours, antibiotic de-escalation occurred in 97% of the patients who met the MCPIS de-escalation criteria.

    The Original and the MCPIS correlated well. For the first evaluator, r=0.99 (p<0.01) for initial and r=0.79 (p<0.01) for 72 hour scores. For the 2nd evaluator, r=0.91 (p<0.01) for initial and r=0.89% (p<0.01) for 72 hour scores. Agreement between the two evaluators was high (Cronbach’s alpha = 0.94 for the initial score and 0.94 at 72 hours).


    Increased efforts are needed to improve sputum collection and antibiotic de-escalation practices.  MCPIS can be easily computed in all patients, correlates well with the Original CPIS, and can serve as a tool for antibiotic de-escalation.  Antibiotic stewardship can play an immense role in helping with antibiotic de-escalation. Further studies are needed to confirm the utility of the MCPIS.

    Nino Balanchivadze, MD1,2, Rajany Dy, MD1,3 and John Boyce, MD, FIDSA1, (1)Yale-New Haven Hospital, New Haven, CT, (2)Grand Strand Medical Center, Myrtle Beach, SC, (3)Pulmonary and Critical Care Medicine, University Of Buffalo, Buffalo, NY


    N. Balanchivadze, None

    R. Dy, None

    J. Boyce, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.