Rapid respiratory panel testing: Impact of active antimicrobial stewardship.

To the Editor:

Polymerase chain reaction (PCR)-based tests are approximately 94% as accurate as viral cultures. The rapid direct respiratory panel (DRP) is a PCR-based diagnostic test that can identify pathogens within hours, compared with standard techniques requiring up to 48-72 hours. This rapid detection can minimize the use of unnecessary antibiotics through timely deescalation when test results do not indicate the presence of a bacterial infection.1, 2, 3 Rapid diagnostic tests are of little value without subsequent antibiotic deescalation when indicated. Antimicrobial stewardship programs (ASPs) can help achieve these outcomes by acting as active messengers and educators for medical teams.4, 5 The purpose of this study was to assess the impact of real-time pharmacy intervention on antibiotic deescalation based on DRP results.

This study was conducted at the Virginia Commonwealth University Health System, an 860-bed tertiary care hospital in Richmond, Virginia. All patients hospitalized between October 2014 and March 2015 were eligible for inclusion in the study. During the study period, an ASP pharmacist reviewed the results of PCR-based DRPs using a computer decision support system (TheraDoc, Charlotte, NC) between 10 AM and 3 PM Monday through Friday, excluding holidays, and paged medical teams with recommendations during these hours. ASP pharmacist recommendations included deescalating antibiotics to complete a course appropriate for community-acquired pneumonia for a subset of these patients when the medical team opted to continue antibiotics.

A total of 68 patients were included in the study. Of these, 55 (80.9%) were receiving antibiotic therapy before the test. Influenza A H3 was the most frequently identified virus (n = 21; 30.9%), followed by rhinovirus, enterovirus (n =16; 23.5%), respiratory syncytial virus (n = 15; 22.10%), influenza A (n = 5; 7.4%), coronavirus OC43 (n = 5; 7.4%), parainfluenza 3 virus (n = 2; 2.9%), parainfluenza 2 virus (n = 1; 1.5%), parainfluenza 4 virus (n = 1; 1.5%), influenza B virus (n = 1; 1.5%), and coronavirus 229E (n = 1; 1.5%). Overall, 66% of the ASP recommendations were accepted. Our ASP pharmacist recommended deescalation or discontinuation of antibiotics in 32% of the cases based on DRP results; however, teams complied with these recommendations in only 19% of cases. The mean interval from the reporting of DRP results to changing antibiotics was 13.5 hours (Table 1 ).

Table 1

ASP recommendations and team actions

ASP recommendation, n (%)	No change: 28 (41.2)
	Discontinue antibiotics: 16 (23.5)
	Decrease duration of antibiotics: 14 (20.6)
	Infectious Diseases consult: 5 (7.4)
	Deescalate therapy: 6 (8.8)
	Escalate therapy: 0 (0.0)
Accepted recommendation, n (%)	Yes: 45 (66.2)
	No: 23 (33.8)
Action following ASP recommendation, n (%)	No change: 40 (58.8)
	Discontinue antibiotics: 6 (8.8)
	Decrease duration of antibiotics: 9 (13.2)
	Infectious Diseases consult: 2 (2.9)
	Deescalate therapy: 7 (10.3)
	Escalate therapy: 4 (5.9)
Time from test result to antibiotic change, h, mean (range)	13.5 (0.5-40)

ASP, antimicrobial stewardship program.

Our study contributes to the growing literature on the impact of real-time ASP interventions based on DRP results. Similar studies have been conducted in the setting of coagulase-negative Staphylococcus and Staphylococcus aureus bacteremia with promising results.6, 7 The results of our study suggest that DRP results tied to direct ASP review and intervention may lead to more prompt antibiotic deescalation, discontinuation, and reduced duration of therapy. Low antibiotic deescalation rates of 10%-20% have been reported in previous studies despite positive DRP results.4, 8 Although 66% of ASP recommendations were accepted overall in our study, only 19% of recommendations to deescalate or discontinue antibiotics were accepted, consistent with these previously published studies.

On the whole, teams were more likely to accept recommendations supporting their current management than those recommending antibiotic deescalation, discontinuation, or Infectious Diseases consultation. The mean time from test resulting in an antibiotic change was 13.5 hours. This highlights a potential opportunity for ASPs to facilitate action sooner based on these data. Further studies are needed to explore the reasons behind failure to deescalate antibiotics despite the availability of DRP results with ASP interpretation. Although rapid diagnostic testing with real-time ASP interpretation and intervention is very promising, the best use of ASP personnel in the reporting and interpretation of these tests remains to be determined.