Funding and Acknowledgments
JMM received a personal 80:20 research grant from Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain, during 2017–19. AD received postdoctoral grants for his stay at the Hospital Clinic of Barcelona from the Lundbeck Foundation and the European Society of Cardiology.
Financial disclosures: JMM has received consulting honoraria and/or research grants from Angelini, Contrafect, Cubist, Genentech, Gilead Sciences, Jansen, Lysovant; Medtronic, MSD, Novartis, Pfizer, and ViiV Healthcare, outside the submitted work. Other authors, no conflicts.
The management of patients with infective endocarditis (IE) has developed markedly during the past several decades. From being a disease treated primarily conservatively with long-term intravenous antibiotics during hospital admission to a disease frequently treated with cardiac surgery, combinations of intravenous and oral antibiotics administered both in and out of hospital.
This chapter is about how to manage the treatment and follow-up of endocarditis patients after hospital discharge. When the decision to discharge an IE patient is made, it is mandatory that the outpatient follow-up plan is in place. At the time of discharge, there are basically three different scenarios. The first scenario is that the patient is considered cured from the infection and therefore terminates the antibiotic treatment before discharge. The second scenario is that the patient is stable enough to allow hospital discharge before the full course of antibiotic treatment is completed and as a result the antibiotic treatment will be finished as outpatient therapy. The third scenario is that the patient is considered incurable due to a combination of age, comorbidity, and severity of IE and therefore discharged to an individualized plan that can span from life-long antibiotics to palliative care (not discussed here).
This chapter will describe the criteria for finishing the antimicrobial therapy at home, either as oral or intravenous administration. In addition, the clinical, microbiological, and surgical follow-up as well as perspectives for future research will be discussed.
Outpatient antimicrobial therapy
There are several benefits of finishing the antimicrobial therapy outside the hospital. Firstly, the patients benefit from being discharged earlier and removed from the stressful hospital environment to the increased comfort in their own homes. Secondly, the patients experience a reduced duration of exposure to hospital acquired infections [ ]. Thirdly, the society profits by reducing the number of hospital admission days and thereby reducing the related health-care costs [ ]. On the other hand, potential downsides of discharging patients before the end of antibiotic treatment must be kept in mind. One possible disadvantage is that the antibiotic treatment might be completed less stringently at home, which in turn could lead to treatment failure. Another challenge is that a latent deterioration of the patient’s condition due to IE might be discovered later when the patient is less frequently assessed at home as compared to in hospital [ ]. While it would be appealing if all patients could finish the treatment in their homes, this is not likely to accomplish and instead the key question is “in which cases will it be safe to finish the treatment out of hospital”? To answer this question, several different criteria have been suggested over time. In the following section, the criteria will be divided in intravenous treatment also called Outpatient Parenteral Antibiotic Therapy (OPAT) and oral treatment, in this chapter named partial oral endocarditis treatment (POET).
Outpatient parenteral antibiotic therapy
OPAT was described as early as around 1980 for treating stable patients with IE or bone, skin, and soft tissue infections [ , ]. During the 1990s, several studies investigated OPAT for patients with streptococcal IE, establishing the evidence for simplifying and reducing the treatment of uncomplicated streptococcal IE [ ]. In 1997, Graninger et al. treated 10 patients with staphylococcal IE with teicoplanin thrice weekly in an outpatient setting and found that 20% needed readmission [ ]. These earlier studies in combination with clinical experience of OPAT led Andrews and Reyn to develop a comprehensive set of criteria to guide the use of OPAT in IE patients [ ]. Due to a few fatal incidents possibly related to OPAT treatment and early discharge, the criteria were created with a relatively selective approach ( Table 13.1 ). Andrews and Reyn defined the initial two weeks treatment as the critical phase where most complications to IE occur and therefore in-hospital treatment was preferred. For the following weeks (2–4 or 2–6 weeks), they suggested that OPAT could be used for medically stable patients that were afebrile, with negative blood culture results and a stable electrocardiogram when changing to OPAT. However, a large part of these patients were not recommended for OPAT based on the following exclusion criteria; aortic valve disease, prosthetic valve disease, acute IE (<2 weeks development), complications to IE (heart failure, abscess, conduction abnormality, mental status change), or IE caused by Staphylococcus aureus, Streptococcus pneumoniae, Haemophilus influenzae, Neisseria meningitidis, Neisseria gonorrhoeae, group B streptococci, gram-negative bacteria or fungi [ ]. These rather restrictive criteria from 2001 to date remain the criteria highlighted in both American and European guidelines on treating IE [ , ]. However, it is important to note that both American and European guidelines recommend modified criteria by not excluding neither specific bacterial etiologies, acute IE onset, aortic nor prosthetic valve disease from receiving OPAT [ , ].
|Critical phase (0–14 days)||The phase were most IE complications occur, and timely diagnosis is important for achieving optimal outcome. |
Preferred management : Inpatient antibiotic therapy for 2 weeks.
Exceptions : OPAT can be considered after 1 week for patients meeting three criteria:
|Continuation phase (Weeks 2–4 or 2–6)||Most patients who have not suffered complications to IE are likely to remain stable during the remainder of therapy, but side effects of parenteral antibiotic therapy may still occur. |
Preferred management : OPAT can be considered for the majority of patients who are medically stable (see point 2 above).
Exceptions : OPAT should generally not be used for patients with any of the following characteristics:
|Essential elements of an OPAT program||Patients should be educated and informed about the complications of IE and indications for and method of contacting their physician or IE care team. Patients and family should be reliable, compliant, and live close to the hospital. Routine postdischarge evaluation should include biweekly office or IE care team home visits during OPAT. Same-day evaluation by a member of the IE care team should be available for patients with recurrent fever or new symptoms.|
Despite the suggested strict criteria, the following years were characterized by case series from many different geographical settings, including a substantial number of patients with prosthetic valve disease, aortic valve disease, and S. aureus etiology for OPAT [ , ]. These findings were summarized in 2019 when Pericàs et al. applied the 2001 criteria to the published OPAT case series and found a general tendency that less than half of the patients fulfilled the 2001 criteria [ ]. In the same paper, they published the results of a multicenter prospective cohort study including 429 patients treated with OPAT thereby successfully reducing the median hospitalization time with 10 days [ ]. This study found the fulfillment of the 2001 criteria unable to distinguish between the patients that actually ended up receiving hospital based antibiotic treatment versus OPAT (fulfillment of criteria in 23.3% vs. 21.7%, P = .465). Instead Pericàs et al. found it reasonable to consider all patients for OPAT when IE was not caused by highly difficult-to-treat microorganisms and there were no clinical, echocardiographic, or postsurgical complications that did not resolve shortly after onset. Based on their national guidelines [ ] and the study they suggested using less strict criteria for OPAT ( Table 13.2 ) [ ]. The suggested new criteria for OPAT included several moderations compared to the 2001 criteria. Firstly, the accepted minimum admission time was shortened from 14 to 10 days in the most uncomplicated cases and the symptom development (acute vs. chronic) was no longer considered a deciding factor. Secondly, the 2019 criteria did not exclude entire species causing IE but instead only excluded highly difficult-to-treat microorganisms requiring intravenous antibiotic combinations that cannot be administered by means of OPAT due to multidrug resistance or potential antibiotic toxicity necessitating close monitoring in hospital. Thirdly, the new criteria did not exclude aortic valve IE and prosthetic valve IE but instead emphasized that a transesophageal echocardiography (TEE) should be performed before changing to OPAT to rule out severe aortic regurgitation and prosthetic valve dysfunction including periannular complications [ ].
|Type of IE||Recommendation||Indications||Requirements|
|Native valve||Rapid shift to OPAT (>10 days after admission/surgery) |
Postponed shift to OPAT (>3 weeks after admission/surgery)
|IE by any causative agent except HDTTM a |
Patients not presenting severe clinical complications
Patients undergoing or not undergoing cardiac surgery
Patients with severe complications at onset
Very fragile patients or patients with severe comorbidities undergoing cardiac surgery or other treatment
|Negative BCs at 72 h |
No severe complications
No anticoagulation issues
TEE ruling out severe aortic regurgitation and prosthetic valve dysfunction.
Same criteria as above plus:
No severe sequelae or clinical complications.
No need for frequent and/or complex cures
|Prosthetic valve||Rapid shift to OPAT (>10 days after admission) |
Postponed shift to OPAT (>3 weeks after admission/surgery)
|IE caused by viridans or bovis group streptococci or Enterococcus faecalis and not undergoing cardiac surgery. |
Patients undergoing cardiac surgery and not infected with HDTTM and without severe complications.
|Same as for rapid shift in NVIE |
Same as for postponed shift in NVIE
|CIED||Rapid shift to OPAT (>7 days after device removal/reimplantation) |
Postponed shift to OPAT (>2 weeks after device removal/reimplantation)
|Cases with no severe clinical complications and not infected by HDTTM. Cases with noncomplicated early lead extraction (<1 week from admission). |
Associated right-sided IE with vegetations > 2 cm.
Left-sided IE, clinical complications, late or complicated lead extraction.
|Normal function of newly implanted device controlled by electrophysiology team. |
No signs of pocket infection
Negative BCs at 72 h after reimplantation and normal TEE
Same criteria as above plus:
No severe sequelae or clinical complications.
No need for frequent and/or complex cures
|Not OPAT candidates||Patients with liver cirrhosis (Child B or C), severe CNS emboli (multiple >3, large >2 cm, hemorrhagic, or with fixed neurologic deficits). Undrained large splenic or renal abscess, vertebral abscesses requiring neurosurgery, periannular complications or other severe conditions requiring surgery when this is contraindicated. b Severe postsurgical complications, HDTTM, and intravenous drug users.|
a HDTTM: Microorganisms requiring IV antibiotic combinations that cannot be administered by means of OPAT or that require strict monitoring of drug levels either in blood or in other fluids owing to their potential toxicity or narrow therapeutic index (e.g., methicillin-resistant Staphylococcus aureus or vancomycin-resistant enterococci also resistant to alternative drugs such as daptomycin and linezolid, multidrug or extensively drug-resistant gram-negative rods, highly penicillin-resistant viridans group streptococci, fungi other than Candida).
It has been estimated that more than quarter of a million people receive OPAT for different infectious diseases every year in the United States [ ]. Despite this, no randomized clinical trials comparing OPAT to in hospital treatment for IE have been performed yet. The different guidelines and criteria are based on expert consensus and observational data regarding complications to IE. Whereas it is not possible to determine which criteria are the best, it seems that the actual clinical practice is closer to the more liberal 2019 criteria than to the 2001 criteria. Naturally, the patients selected for OPAT treatment are generally in a better condition than those patients that will have to finish their treatment in hospital. Therefore, despite the relatively low mortality rates in OPAT observational studies, we are currently unable to know if it would have been even lower if they were treated in-hospital. Until further clinical evidence exist, it seems that the recipe for successful OPAT treatment is the rather open interpretation in the international guidelines (US and European) in combination with extensive clinical experience in the IE teams treating these patients.
In addition to selecting the correct patients for OPAT treatment, it is crucial to have the necessary logistic setup to perform the OPAT. The required setup includes patient education and a well-organized multidisciplinary team. Patients should be thoroughly informed about IE complications and symptoms of treatment failure and how to contact the IE team for a same-day evaluation [ ]. The OPAT administration should be secured by daily visits by a skilled IE nurse also checking for complications including catheter infections. At least twice weekly, the patient should attend the outpatient clinic or receive a visit by the attending physician to check the general progress in the patient’s condition and to exclude signs of new IE complications [ , ].
Partial oral endocarditis treatment
POET is defined as changing to oral antibiotics for the latter part of the endocarditis treatment. In principle, the oral treatment can be administered in hospital, but the key objective with the POET is to discharge the patients earlier to obtain the advantages mentioned above in the OPAT section. Additional advantages of POET compared to OPAT are that there is no need for medical staff to help administer the IV medicine and the reduced time with an IV line in situ, leads to a reduced exposure to catheter-associated infections [ ].
The concept of POET treatment was described in several studies in the 1980–1990s in the treatment of streptococcal and staphylococcal IE [ , ]. Parker et al. cured 35 cases of staphylococcal IE with a mean of 16 days IV antibiotics followed by a mean of 26 days oral treatment with primarily dicloxacillin or oxacillin [ ]. With more questionable effect, two smaller observational studies (15 and 11 patients) treated uncomplicated left-sided native valve streptococcal IE with 6 weeks of oral ampicillin, respectively, as mono therapy and in combination with 2 weeks of intramuscular streptomycin [ , ]. Whereas the direct oral strategy seemed too risky, Stamboulian et al. cured all streptococcal IE patients in a randomized clinical trial comparing 15 patients treated with 4 weeks IV ceftriaxone versus 15 patients treated with 2 weeks IV ceftriaxone followed by 2 weeks oral ampicillin [ ]. Dworkin et al. cured 10 out of 13 IV drug users with right-sided native valve S. aureus IE with 1 week of IV ciprofloxacin followed by 3 weeks oral ciprofloxacin in combination with 4 weeks of oral rifampicin [ ]. The same pattern was found in 1996 when Heldman et al. successfully completed a randomized clinical trial in IV drug users with right-sided S. aureus IE, treating 19 patients with 4 weeks of oral ciprofloxacin and rifampicin with the same efficacy as 25 patients treated with 4 weeks IV oxacillin or vancomycin in combination with gentamicin [ ]. More than a decade later in 2011, Demonchy et al. switched 19 patients (12 with staphylococcal IE and 7 with streptococcal IE) to oral treatment after a mean of 18 days of IV antibiotics and found that none of the patients died [ ]. The largest retrospective study so far was published in 2016 and included 426 IE patients with different bacteriology (primarily streptococci, staphylococci, and enterococci) and all types of IE (right- and left-sided + native valves and prosthetic valves + cardiac implantable electronic devices) [ ]. From the total retrospective cohort, 214 patients were switched to oral treatment after a median of 21 days of IV antibiotic treatment and in a multivariate analysis the authors did not find a switch to oral antibiotics to be an independent risk factor of adverse outcome.
Finally, in 2018 Iversen et al. published the results of the long expected prospective, randomized multicenter POET trial including 400 patients with left-sided IE to receive either standard IV treatment for the entire course or switch to oral treatment during the course [ ]. They included stabilized patients with a satisfactory clinical response to initial IV antibiotic treatment (minimum 10 days and at least 7 days after cardiac surgery) suffering from IE caused by streptococci, Enterococcus faecalis , S. aureus , or coagulase-negative staphylococci ( Table 13.3 ). The study was designed to show noninferiority of partial oral treatment with a composite primary endpoint of all-cause mortality, unplanned cardiac surgery, symptomatic embolic event, and relapse of bacteremia (<6 months). The primary endpoint was predicted to occur in 10% (5%–13%) of the patients and the power calculation was based on an absolute risk difference of 10% points (relative risk difference of 100%) as the clinically relevant difference used as the noninferiority margin. The composite endpoint occurred in 12.1% in the IV group and in 9.0% in the orally treated group, resulting in an absolute difference between groups of 3.1 percentage points (95% CI, −3.4 to 9.6; P = .40) meeting the noninferiority goal. Consequently, the results turned out satisfactory for the oral treatment with an acceptable risk of missing an absolute difference of 3.4% points (relative risk difference around 30%) in favor of IV treatment. Regarding safety, adverse events from antibiotics were similar in the two groups (6% vs. 5%). A major strength of the POET trial was the strict inclusion criteria that accordingly can be used as a clinical guide to elect patients for the switch to oral treatment during the IE treatment course ( Table 13.3 ).