In a cardiology department, there are some patients that require long-term antibiotics, such as those with infective endocarditis or infected prosthetic devices. We describe our experience with intravenous antibiotic therapy for patients with cardiology diagnoses who require a period of antibiotics in our outpatient service during the period of the COVID-19 pandemic. A total of 15 patients were selected to have outpatient antibiotic therapy (age range 36 to 97 years, 60% male). A total of nine patients had infective endocarditis, four patients had infected valve prosthesis or transcatheter aortic valve implantation (TAVI) endocarditis, one patient had infected pericardial effusion while another had infected pericarditis. For these 15 patients there was a total of 333 hospital bed-days, on average 22 days per patient. These patients also had a total of 312 days of outpatient antibiotic therapy, which was an average of 21 days per patient. The total cost, if patients were admitted for those days, assuming a night cost £400, was £124,800, which was on average £8,320 per patient. Three patients were readmitted within 30 days. One had ongoing endocarditis that was managed medically and another had pulmonary embolism. The last patient had a side effect related to daptomycin use. In conclusion, outpatient antibiotic therapy in selected patients with native or prosthetic infective endocarditis appears to be safe for a selected group of patients with associated cost savings.
A major contributor to the cost of a patient’s care is the number of days occupying hospital beds. In a cardiology department, there are some patients that require long-term antibiotics, such as those with infective endocarditis or infected prosthetic devices. While most of these high-risk patients require some duration of monitoring for complications and deterioration as inpatients, there are patients who may, after a period of observation, be stable enough to be discharged home with outpatient intravenous therapy. In this report, we describe our experience with intravenous antibiotic therapy for patients with cardiology diagnoses who required a period of antibiotics in our outpatient service during the period of the COVID-19 pandemic.
Method and materials
We included patients discharged between January 2020 and September 2020 who had treatment plans to receive outpatient parenteral antimicrobial therapy (OPAT). Patients were identified via the nurse-led OPAT service. Our trust follows the local guidelines, which are based on the OPAT best practices,1 and patients were considered appropriate for OPAT referral based on the infection specialist’s decision. The OPAT service consists of three infectious disease consultants, one consultant microbiologist, two antimicrobial pharmacists and 2.4 clinical nurse specialists. The OPAT clinical nurse specialists assess the patient while they are in hospital to ascertain if they meet the OPAT criteria. They also liaise with consultants and microbiology to instigate an appropriate treatment plan, consent patients to the service and insert or arrange appropriate intravenous access. Once the patient is discharged, the OPAT nurses review patients in clinic and monitor the bloods, liaise with other specialties and chair the weekly OPAT multi-disciplinary meeting. The actual process of administering the antibiotics at home is undertaken by the local district nursing teams, unless the patient is capable of self-administration. Self-administration involves a six-hour training programme, which is delivered by the OPAT nurses. Occasionally, patients will attend the OPAT clinic daily to receive their antibiotics, but this is not feasible for patients on long-term treatment.
Data were collected from electronic records on the age, sex, presentation, past medical history, investigation results, diagnosis and outcomes for patients. Investigations of interest were organisms isolated from blood cultures, echocardiogram findings and relevant imaging. The treatments patients received including type of outpatient antibiotic, duration of treatment and any readmissions or adverse outcomes were collected.
A total of 15 patients were selected to have outpatient antibiotic therapy (table 1). The age of these patients ranged from 36 to 97 years and 60% were male. Most patients presented to hospital feeling generally unwell with other symptoms, such as shortness of breath, rigors, chest pain, cough, confusion and sweats. A total of nine patients had infective endocarditis, four patients had infected valve prosthesis or transcatheter aortic valve implantation (TAVI) endocarditis, one patient had infected pericardial effusion while another had infected pericarditis. All but one patient had multiple cardiovascular and non-cardiovascular comorbidities. There were six patients with prosthetic valves and one patient had an early infection while the other five patients had late infections. Blood cultures isolated a variety of organisms including Staphylococcus and Streptococcus species. The most common treatment was inpatient antibiotics followed by a period of outpatient antibiotics, which was typically once or twice daily ceftriaxone 2 g. Other treatment regimens included benzylpenicillin or flucloxacillin as continuous infusions, daptamicin and rifampicin and a combination of ceftriaxone and teicoplanin. Three patients were readmitted within 30 days. One had ongoing endocarditis that was managed medically and another had pulmonary embolism. The last patient had a side effect related to daptomycin use.
Table 1. Patient characteristics, diagnosis, treatment and outcomes
|1||72||F||Native mitral valve endocarditis||Surgery with tissue mitral valve replacement. IP ABX and OP ceftriaxone 2 g BD for 15 days||Readmissions with unwell and raised CRP|
|2||97||M||Native valve infective endocarditis||IP ABX and 12 days of OP ceftriaxone 2 g OD||Readmission with pulmonary embolism|
|3||78||F||Native tricuspid endocarditis||IP ABX and 16 days of ceftriaxone 2 g BD||None|
|4||47||F||Pericarditis||IP ABX and 18 days of ceftriaxone 2 g OD and teicoplanin 80 mg||None|
|5||68||F||Late prosthetic aortic valve endocarditis||IP ABX and 12 days of ceftriaxone 2 g OD||None|
|6||36||M||Infected pericardial effusion||IP ABX and 36 days of ceftriaxone 2 g OD||None|
|7||66||M||Native aortic and pulmonary valve endocarditis||IP ABX and 20 days of ceftriaxone 2 g BD||None|
|8||52||M||Native aortic valve endocarditis||Revision of total hip replacement. IP ABX and 28 days of ceftriaxone 2 g BD||None|
|9||46||F||Early prosthetic valve endocarditis||IP ABX and 13 days of ceftriaxone 2 g BD||None|
|10||85||F||Late prosthetic aortic valve endocarditis||IP ABX and 14 days for daptomycin 350 mg and rifampicin||Readmission with elevated CK. Daptomycin switched to flucloxacillin 4 g/day|
|11||75||M||Late prosthetic aortic valve endocarditis||IP ABX and 29 days of OP ceftriaxone 2 g OD||None|
|12||69||F||Native mitral valve endocarditis||IP ABX and OP flucloxacillin 12 g/day for 22 days||None|
|13||64||M||Native mitral valve endocarditis||IP ABX and OP benzylpenicillin 7.2 g/day for 19 days||None|
|14||84||M||Late prosthetic tissue aortic valve endocarditis||IP ABX and had OP ceftriaxone 2 g OD for 30 days||None|
|15||86||M||Suspected late prosthetic tissue aortic valve endocarditis||IP ABX and OP ceftriaxone 2 g OD for 28 days before admission with shortness of breath||None|
|Key: ABX = antibiotics; BD = twice daily; CK = creatine kinase; CRP = C-reactive protein; F = female; IP = inpatient; M = male;
OD = once daily; OP = outpatient
Table 2. Length of stay and treatment and cost savings
|Total inpatient days for initial treatment||333 days|
|Average days per patient||22 days|
|Total outpatient days for antibiotics||312 days|
|Average days of outpatient antibiotics per patient||21 days|
|Total days for initial treatment and outpatient antibiotics||645 days|
|Savings from inpatient bed days if cost was £400/night||£124,800|
|Savings from inpatient bed days per patient||£8,320|
Table 2 shows the length of stay, treatment and associated cost savings. For these 15 patients there was a total of 333 hospital bed-days, on average 22 days per patient. These patients also had a total of 312 days of outpatient antibiotic therapy, which was an average of 21 days per patient (range six to 36 days). The total cost, if patients were admitted for those days, assuming a night cost of £400, was £124,800, which was on average £8,320 per patient. The average cost for each drug regimen is shown in table 3.
These findings suggest that OPAT in cardiology patients can be associated with significant cost savings without major adverse outcomes for patients. While we observed that 20% of patients returned to hospital, it is noted that these patients are a high-risk group. However, once they are initiated on antibiotics and found to be stable, some can be selected to go home and have outpatient intravenous antibiotic therapy. Keeping patients in hospital is not only associated with a high cost, due to inpatient bed costs, but also places patients at risk of hospital-acquired infections, such as pneumonias and Clostridium difficile, which can be a major problem when patients are on antibiotics.
Table 3. Average drug costs per day per regimen
|Medicine||Regimen||Cost per day*|
|Ceftriaxone vials for injection||2 g OD||£19.16|
|Ceftriaxone vials for injection||2 g BD||£38.32|
|Teicoplanin vials for injection||800 mg OD||£14.64|
|Daptomycin + rifampicin||IV 350 mg OD + PO 600 mg BD||£62.00 + £2.46|
|Flucloxacillin vials for injection||4 g/day IV||£12.00|
|Flucloxacillin elastomeric device||12 g/day||Not stated|
|Benzylpenicillin elastomeric device||7.2 g/day||Not stated|
|*Prices from British National Formulary July 2021
Key: BD = twice daily; IV = intravenous; OD = once daily; PO = per oral
An important aspect of an OPAT service is patient selection. A study of 80 patients with infective endocarditis from a single centre over 12 years found that patients with cardiac failure, renal failure or teicoplanin therapy were at increased odds of OPAT failure, defined by readmission or switch of antibiotics.2 Using the same definition for failure, the rate of failure in this study was 26% compared with the 31.3% reported in the study, as three patients were readmitted and one patient had changed from daptomycin to flucloxacillin because of elevated creatinine kinase. A previous review suggests that patients with native valve endocarditis should have at least two weeks of treatment in hospital where complications are most likely to occur and consideration for OPAT for two or four more weeks when life-threatening complications are less likely.3 In a Spanish study, the safety of outpatient antibiotics has also been demonstrated for 392 patients with uncomplicated viridans-group Streptococci infective endocarditis of which 16% of patients were readmitted while 4% died.4 In the UK, a retrospective review of 36 patients with infective endocarditis found that outpatient therapy is safe for infective endocarditis as well as prosthetic valve endocarditis and other cases at higher risk of complicated disease. Rates of adverse events occurred in a third of patients of which more than half were associated with lines.5 We have built on these findings by considering the cost benefit of outpatient therapy, which is considerable. Furthermore, it appears to be a safe alternative compared with a prolonged inpatient admission, where patients could be potentially at risk of contracting COVID-19, and may have associated improvements in patient satisfaction.
A recent trial of 400 patients with left-sided endocarditis, randomised antimicrobial therapy to intravenous or oral routes after initial treatment. It was found that changing to oral antibiotics was non-inferior to intravenous antibiotics.6 The approach of oral antibiotics may be associated with further saving as the cost associated with the nurse administering therapy to patients can be reduced. In a low-risk population, it is likely that oral antibiotic treatment is safe. However, in a high-risk population, the nurse visit is an opportunity to review these patients for deterioration and complications, which is not possible if the patient is discharged with self-administered oral antibiotic therapy. In our current evaluation, there was no mortality in the cohort and tissue mitral valve replacement was urgently performed in one patient who required surgery. A larger comparative study is needed to better understand the safety of part-home delivered antibiotic therapy for infective endocarditis.
In conclusion, OPAT in selected patients with native or prosthetic infective endocarditis appears to be safe for a selected group of patients. The benefits above safe patient care include a major saving in costs, which can be more than £8,000 on average per patient.
Conflicts of interest
This work was conducted as a clinical audit and health service evaluation. Following trust policies, it was registered with the audit department and was assigned a reference number of CA31221.
Supplementary tables providing more patient and microbiological details are available on request from the author.
1. Chapman ALN, Patel S, Horner C et al. Updated good practice recommendations for outpatient parenteral antimicrobial therapy (OPAT) in adults and children in the UK. JAC Antimicrob Resist 2019;1:dlz026. https://doi.org/10.1093/jacamr/dlz026
2. Duncan CJA, Barr DA, Ho A et al. Risk factors for failure of outpatient parenteral antibiotic therapy (OPAT) in infective endocarditis. J Antimicrob Chemother 2013;68:1650–4. https://doi.org/10.1093/jac/dkt046
3. Andrews MM, von Reyn CF. Patient selection criteria and management guidelines for outpatient parenteral antibiotic therapy for native valve infective endocarditis. Clin Infect Dis 2001;33:203–09. https://doi.org/10.1086/321814
4. Cervera C, del Rio A, Garcia L et al. Efficacy and safety of outpatient parenteral antibiotic therapy for infective endocarditis: a ten-year prospective study. Enferm Infecc Microbiol Clin 2011;29:587–92. https://doi.org/10.1016/j.eimc.2011.05.007
5. Partridge DG, O’Brien E, Chapman ALN. Outpatient parenteral antibiotic therapy for infective endocarditis: a review of 4 years’ experience at a UK centre. Postgrad Med J 2012;88:377–81. https://doi.org/10.1136/postgradmedj-2011-130355
6. Iversen K, Ihlemann N, Gill SU et al. Partial oral versus intravenous antibiotic treatment of endocarditis. N Engl J Med 2019;380:415–24. https://doi.org/10.1056/NEJMoa1808312