Outpatient parenteral antimicrobial therapy (OPAT) involves intravenous (IV) administration of antibiotics to patients whose treatments can be safely managed at their own homes (Paladino and Poretz, 2010; Seaton and Barr, 2013). This usually involves nurses visiting patients' homes on a regular basis. Multiple tasks are performed during each visit, including delivery of antibiotics, collection of blood samples and provision of care for their IV devices, as well as clinical assessment of their progress.
The number of daily visits required varies among OPAT patients, depending on their conditions as well as the type of antimicrobial treatment. Certain antimicrobial treatments, such as piperacillin/tazobactam (tazocin) and flucloxacillin, require more doses per day and, hence, more daily visits than broad-spectrum (i.e. target-nonspecific) antibiotic treatments, such as ceftriaxone, which requires only one dose per day. For example, the standard administration of IV tazocin involves three doses administered at regular intervals every day, requiring multiple home visits and putting a considerable amount of pressure on the limited resources of local OPAT teams. This is particularly a concern for antimicrobial treatments that require even more frequent doses per day. In the authors' experience, as an example, none of the community teams in Oxfordshire, England, have the capacity to provide out-of-hours IV flucloxacillin administration at patients' homes, which involves four doses at regular intervals per day. Hence, patients requiring flucloxacillin either stay in hospitals for IV treatments or receive broad-spectrum antibiotic treatments that require fewer doses per day. These alternative options are suboptimal for the patient.
The issues of limited resources and suboptimal drug choices can be mitigated using infusion devices, such as elastomeric pumps, that allow drug delivery at a constant, predetermined rate across the entire day (Thompson et al, 2011; Allwood et al, 2020). Notably, in the climate of the COVID-19 pandemic, the use of infusion devices can reduce home visits, thus minimising transmission of COVID-19 to OPAT patients who are at a high risk of COVID-19-triggered complications. Since 2019, the authors' team at the John Radcliffe Hospital, Oxford, has started to use the infusion device Accufuser (Vygon, UK) for patients who require IV flucloxacillin treatments. Although previous research has investigated and confirmed the safety of Accufuser among outpatients (Oliver, 2016), the clinical effectiveness of these infusion devices relative to standard antibiotic treatments has yet to be determined.
In this retrospective study, the clinical effectiveness of Accufuser infusion of flucloxacillin was explored in a small subpopulation of OPAT patients (n=16), by comparing the total number of days these patients spent under OPAT care and their clinical outcomes at the time of discharge with patients having the same diagnosis but receiving standard antimicrobial treatments (n=74). As patients were medically stable at the start of OPAT, and the total number of days under OPAT care was used as an indirect measure of the clinical effectiveness of the antibiotic treatment: more prolonged OPAT care indicated that the treatment was clinically less effective. Further, at the time of discharge, patients received a score of their clinical outcome. An OPAT score of 1 indicates that the OPAT treatment has been successful, and readmission to the hospital is not required, while a score of 0 indicates that readmission to the hospital is necessary. The OPAT scores of patients receiving Accufuser infusion were compared with those of patients having the same diagnosis but receiving standard treatment.
Methods
Data availability
Data available are part of the routine record of all discharged cases from the ambulatory outreach team (AOT) at the John Radcliffe Hospital, which provides OPAT services across Oxfordshire, England. These discharge data have been fully anonymised, uploaded and shared publicly on the OPAT National Outcomes Registry System (NORS) (British Society for Antimicrobial Chemotherapy, 2017). Anonymised records included in the present study were from the second quarter of 2019 to the first quarter of 2020, with patients diagnosed with three primary infections: bloodstream infections (BSIs), non-vertebral osteomyelitis and vertebral osteomyelitis.
Data analyses
Statistical analyses were conducted using SPSS version 25. One-way between-subjects analyses of variance (ANOVAs) were conducted on the numbers of days under care. Binomial and sign tests were conducted on the number of successful outcomes among patients receiving Accufuser infusion and those receiving standard antibiotic treatments. α=.05 was adopted in all statistical analyses.
Results
Days under OPAT care
The number of days a patient spends under OPAT care varies greatly, depending on the patient's primary infective diagnosis, as OPAT patients represent a heterogenous clinical population. In this paper, the days under OPAT care were compared between Accufuser infusion and standard methods of antibiotic administration. The first analysis examined if there was any beneficial effect of Accufuser delivery over and above standard methods of antibiotic administration among patients with three BSIs, non-vertebral osteomyelitis and vertebral osteomyelitis.
Table 1 lists the number of patients with each infective diagnosis (BSIs, non-vertebral osteomyelitis or vertebral osteomyelitis) and treatment protocol (standard or Accufuser antibiotic therapy), along with a comparison of days under OPAT care for each of these conditions.
Table 1. Mean numbers of days (± standard errors of the mean) under OPAT care
| Diagnosis | Treatment | N | Mean days | SEM | F | Significance |
|---|---|---|---|---|---|---|
| Bloodstream infection | Standard | 52 | 17.70 | 4.90 | .01 | p=.93 |
| Accufuser | 10 | 16.80 | 1.85 | |||
| Non-vertebral osteomyelitis | Standard | 8 | 28.38 | 4.71 | 6.74 | *p<.05 |
| Accufuser | 4 | 9.75 | 3.30 | |||
| Vertebral osteomyelitis | Standard | 14 | 16.93 | 2.60 | 7.26 | *p<.05 |
| Accufuser | 2 | 36.50 | 5.50 |
Note: OPAT=outpatient parenteral antibiotic therapy; SEM=standard error of the mean, calculated from the standard deviation divided by the square root of N in each condition; F=omnibus F from one-way analysis of variance (ANOVA)
Separate one-way, between-subjects ANOVAs (Accufuser infusion vs. standard administration) were performed, one for each of the three patient subgroups. Among patients with BSIs, there was no significant main effect of antibiotic delivery (mean days±SEM in the standard vs. accufuser conditions=17.70±4.90 vs. 16.80±1.85; F(1,60)=.01, p=.93; Table 1). However, among patients with non-vertebral osteomyelitis patients, there was a significant main effect of antibiotic delivery [F(1,10)=6.74, p<.05], with those receiving Accufuser infusion spending fewer days under OPAT care than patients receiving standard antibiotic administration (mean days ± SEM in the standard vs. Accufuser conditions=28.38±4.71 vs. 9.75±3.30; Table 1). In contrast, among patients with vertebral osteomyelitis, the exact opposite pattern was observed, with those receiving Accufuser infusion spending more days under OPAT care than patients receiving standard antibiotic administration (mean days ± SEM in the standard vs. Accufuser conditions=16.93±2.60 vs. 36.50±5.50; F(1,14)=7.26, p<.05; Table 1). Thus, there was a complex relationship between the method of antibiotic delivery and the patient's condition, and the beneficial effect of Accufuser infusion in a certain patient population may not be easily generalised.
Clinical endpoints
At the time of discharge, each patient received a score of their clinical outcome. To assess any potential effect of Accufuser delivery of antibiotics on patients' outcomes at the end of care, binomial tests were conducted on the number of successes (which were defined as cases with an OPAT score of 1) among patients receiving Accufuser infusion (n=16) and among those receiving standard antibiotic administration (n=74), pooling data across the three infections. The number of successes was then tested against the assumption that there is an equal probability of success and failure (i.e. the sign test). For both Accufuser infusion and standard delivery conditions, observed probabilities of success (94% and 86%, respectively) were significantly greater than chance level of 50% (ps<.005). In addition, the number of successes in the Accufuser condition was not significantly different from the observed probability of success for the standard condition (binomial test, p=.43). Taken together, these findings suggest that, although Accufuser infusion may affect discharge rates (as measured by the number of days under OPAT), this method of antibiotic delivery does not affect the patient's clinical outcome at the end of treatment.
Discussion
In the present study, the clinical effectiveness of delivering flucloxacillin using the Accufuser infusion devices was explored in a small sample of OPAT patients. It is more cost effective than standard methods of antibiotic administration, as patients receiving Accufuser infusion require only one home visit per day rather than the standard four doses per day (i.e. a 75% reduction in contact frequency). Crucially, in the present climate, reducing home visits can also help to minimise transmission of COVID-19 to OPAT patients who are at a high risk of COVID-19-triggered complications. As patients are medically stable at the start of OPAT, the number of days under OPAT care is used as an indirect measure of the clinical effectiveness of the antibiotic treatment: more prolonged OPAT care indicates that the treatment is clinically less effective.
The novel finding of this study is that there is a complex relationship between the method of antibiotic delivery and the patient's condition. In patients with BSIs, the clinical effectiveness of an Accufuser pump did not differ from that of standard administration, as indicated by the similar numbers of days under OPAT care (Table 1). Thus, for this group of patients, Accufuser infusion is a better option, in terms of both costs and patient contact and therefore transmission of diseases from hospitals to vulnerable populations. Among patients with non-vertebral osteomyelitis, those receiving Accufuser infusion spent fewer days under OPAT care than those receiving standard antibiotic administration (Table 1), indicating multiple advantages of using the Accufuser pump over standard treatments (e.g. in terms of clinical effectiveness, costs and patient contact). However, among patients with vertebral osteomyelitis, those receiving Accufuser infusion spent more days under OPAT care (i.e. slower to be discharged) than those receiving standard antibiotic administration (Table 1). Thus, there might be adverse clinical effects of using Accufuser pumps in this group of patients.
It is unclear why the clinical effectiveness of Accufuser antibiotic infusion varies among OPAT patients with different infective diagnoses. Anecdotal evidence from the authors' team suggested that antibiotic residues are occasionally found inside Accufuser pumps. Thus, any potentially adverse effects of Accufuser pumps observed in patients with vertebral osteomyelitis could be, at least in part, due to incomplete antibiotic delivery. Crucially, vertebral osteomyelitis patients might be more susceptible to incomplete antibiotic treatment than the other two subgroups, which would prolong the course of OPAT care.
Interestingly, in the authors' experience, incomplete antibiotic delivery tends to be more frequent during the winter. Although the cause of seasonal variation in the number of residue-related incidents is unclear, the performance of Accufuser pumps is likely to be sensitive to ambient temperature, an issue that has been reported in a different type of elastomeric pumps (Ehrmann et al, 2004). These pressing issues must be addressed in order to gain a more comprehensive understanding of the differential clinical effectiveness of Accufuser infusion among OPAT patients with different infective diagnoses.
Conclusion
The preliminary findings of this study suggest that there is a complex relationship between the method of antibiotic delivery and the patient's condition, and the benefits of Accufuser infusion in a certain OPAT patient population may not be easily generalised. It should be emphasised that sample sizes in the two osteomyelitis groups that received Accufuser infusion were small in this retrospective study. These statistically significant effects were exploratory in nature, and there are other potential confounders (e.g. age, length of hospital stay prior to OPAT treatment and other comorbidities) that may have influenced the results. Given the interesting results of this study, as well as the benefits and growing popularity of using infusion devices among outpatients, further studies examining the clinical effectiveness of Accufuser infusion with larger cohorts of outpatients are urgently needed.
KEY POINTS
- Using elastomeric infusion devices is a cost-effective way to deliver outpatient parenteral antimicrobial therapy (OPAT), and it reduces patient contact and, hence, minimises COVID-19 transmission to outpatients
- The clinical effectiveness of antibiotic administration using elastomeric infusion devices may vary among patients with different infective diagnoses
- Given the surging interest in using elastomeric infusion devices for outpatients, further studies with larger samples are needed to examine their clinical effectiveness in outpatients with different infective diagnoses
- There is a complex relationship between the method of antibiotic delivery and the patient's condition
CPD REFLECTIVE
- How can a balance be struck between the cost-effectiveness and clinical effectiveness of elastomeric infusion devices?
- What are some factors that affect the efficacy of antibiotic administration in outpatient settings?
- What are the possible reasons why the clinical efficacy of antibiotic administration using elastomeric infusion devices differs depending on the patient's diagnosis?