These data show that the use of RPCR instead of standard MPCR is associated with benefits in antimicrobial stewardship through reduction in unnecessary antibiotic prescribing as well as the provision of oseltamivir therapy for people at high risk of complications with confirmed influenza. Taken together these data show a very positive impact on appropriate antimicrobial prescribing following introduction of a RPCR influenza test at our institution.
Antibiotic prescription significantly decreased from 67% to 51% when RPCR was used indicating a decrease in unnecessary antibiotic prescription. It is well recognised that antimicrobial resistance is a serious global threat to health of which unnecessary antibiotic use is a contributor. A key objective of the World Health Organisation (WHO) Global Action Plan on Antimicrobial Resistance is to optimise the use of antimicrobial medications recommending effective and rapid diagnostic tools as part of this plan. In line with WHO guidelines, our results would support the use of RPCR as opposed to standard MPCR as an effective diagnostic tool to improve antimicrobial stewardship. Rapid testing has previously been suggested as a means to improve antibiotic utilisation however these studies have been completed using rapid antigen test which is has lower sensitivity and were studies performed in outpatient ED settings, paediatrics or resource limited regions.[11, 30–33] Our findings extend on and are supported by a small number of previous studies that suggested a reduction in antibiotic usage with RPCR testing.[14, 17–19, 23] In comparison to our study these were are mostly small studies assessing effects on ED metrics and included non-comparative cohort studies [14–16] or used historical controls[17–23] In particular, one trial found antimicrobial stewardship improvement only occurred in their paediatric and not in their adult population. Notably there has been one large, UK hospital-based, open-label randomised controlled trial comparing outcomes of routine RPCR at presentation of respiratory illness versus standard clinical care. In this UK study, those allocated to the standard clinical care group were provided treatment at the discretion of treating teams including potential conventional respiratory viral PCR testing as well as antimicrobial use. This study found that routine RPCR did not reduce duration of antibiotics overall, but did find that patients in the RPCR group received either single dose or shorter courses of antibiotics compared to the control group. Though they found higher proportions of single dose/shorter courses of antibiotics in their RPCR group, the effect appeared to have been muted in the overall duration of antibiotics due to clinical populations in their study, such as pneumonia, requiring consistently longer courses of antibiotics. Our findings would suggest that earlier diagnosis could alter clinical management such that unnecessary antibiotic use can be decreased. In reducing unnecessary antibiotics, not only will this reduce antimicrobial resistance, but patients and hospitals could thereby reduce known risks associated with antibiotic overuse such as antibiotic associated adverse effects, re-attendance due to infectious disease, increased healthcare costs or increased length of stay.
Subgroup analyses of higher risk patients for complications of influenza would suggest that RPCR testing lead to fewer missed opportunities for oseltamivir treatment compared to MPCR testing (24% vs 37%, respectively). Early oseltamivir use has benefits in improving resolution of clinical symptoms, reducing risk of lower respiratory tract infections and prevention of hospital admission when used in influenza positive patients. Current Australian Therapeutic Guidelines recommend considering treatment for individuals at risk of poor outcomes such as elderly, pregnant women, immunocompromised, etc. A recent randomised controlled trial demonstrated oseltamivir treatment in the older, co-morbid population can improve recovery by two to three days, further supporting the need to target early therapy especially in these higher risk individuals. This is the only study to our knowledge that has looked at the use of RPCR diagnostic testing to optimise oseltamivir use in high risk individuals for complications of disease. A small, single centre US study comparing MPCR to RPCR, found that negative RPCR results lead to reduced duration of empiric antiviral therapy with no difference in intensive care admissions or antibiotic use. . This study, however, examined both positive and negative influenza results and those that were positive only accounted for 11% of the study population. Small non-comparative studies, within the ED setting alone, have also suggested that utilisation of RPCR may improve oseltamivir prescription.[15, 16, 21, 36] Both our study, in conjunction with the US study and non-comparative studies would support the idea that a RPCR test compared to a MPCR test improves the utilisation of oseltamivir.
Utilisation of either diagnostic test did not appear to affect LOS in the ED or as an inpatient, as neither result achieved statistical significance. Logistical factors occurring within the ED or inpatient wards that were not assessed within this study may have played a role. There have been a small number of studies published on RPCR testing and its impact on LOS both in ED and inpatient settings.[34, 37–39] Overall, these studies have been of varying design and have demonstrated different outcomes. Factors identified that affected LOS included being provided positive RPCR results during hospital stay or delays in timing of patient diagnostic sampling.[38, 39] Our study did not examine time difference of patient presentation to sampling nor had we analysed patient disposition at time of result availability which may have impacted LOS and contributed to our findings. The present study had a smaller sample size and may not have sufficient power to detect a difference in LOS.
This study’s limitations include that data were from a single hospital, were collected retrospectively and could have been subject to selection bias at the time of test selection (MPCR vs RPCR). However, to try and account for this we checked to ensure that the groups did not differ significantly on the basis of age, mortality or morbidity. Data were collected through reviewing each individual’s medical record, and therefore relied on the accuracy of documentation at time of presentation. Medical history and clinical data were extracted from the medical records, therefore, to minimise bias, the same author collected all the data reducing variability in interpretation. Population numbers in the RPCR group were lower than the multiplex group, likely reflecting lower utilisation after introduction of a new testing method.
Future studies could utilise larger, prospective, randomised trials with direct comparison of standard multiplex and rapid PCR testing to monitor outcomes. Larger uptake in test utilisation and influenza positive cases may improve the statistical power and would be of interest in comparison to this study. Research questions that could be further explored in a prospective randomised trial include effects on bed management, patient flow and an exploration of the multi-level factors that impact LOS. Cost-benefit analyses was not assessed in this study. We note that the RPCR is significantly more expensive than the standard MPCR with an expected cost of approximately AUD$50 per test. Future studies could examine the cost-benefit of the RPCR test.