Surgery of the Left Heart Valve Infective Endocarditis
Keywords: Aortic Valve Repair, Operations for Valvular Heart Disease, infective endocarditis native valve endocarditis, prosthetic valve endocarditis
· Infective endocarditis (IE) is the most severe and devastating complication of heart valve disease, whether it is native valve endocarditis (NVE), prosthetic valve endocarditis (PVE), or infection on another cardiac device. Despite advances in surgical technique, operations for IE remain associated with the highest mortality of any valve disease.
· IE patients require a multispecialty team approach, which includes an infectious disease specialist, cardiologist, and cardiac surgeon, with input from other specialties, such as neurology and nephrology, when needed. This is because the clinical scenarios presented by patients with IE are often very complex and require prompt diagnosis for the early institution of antibiotic treatment and decision making related to complications, including the risk of embolism and need for and timing of high-risk surgery.
· The microbiology of IE depends on whether the valve is native or prosthetic and whether the infection is community or hospital acquired. Staphylococci, streptococci, and enterococci are responsible for about 85% of all cases of IE.
· The infecting organisms produce and release virulence factors, including toxins, and enzymes. The enzymes produced are organism-specific regarding tissue specificity and efficiency. The severity of invasion and destruction, involvement of the valve annulus and beyond, occurs in stages cellulitis, abscess, abscess cavity, and finally pseudoaneurysm and are a function of virulence and time, with Staphylococcus aureus being the most aggressive and destructive.6
· The capacity of biofilm production, which protects bacteria from host immune defenses and impedes antimicrobial efficacy, thus significantly reducing the ability of medical therapy alone to eradicate the infection, is a hallmark of microorganisms commonly causing IE.5
· A high index of suspicion and low threshold to perform the examination and studies necessary to exclude IE are essential to diagnosis and early treatment. The diagnosis of IE is based on clinical symptoms, physical findings, microbiology results, echocardiograms, and other results. Echocardiography and blood cultures are the cornerstones of diagnosing IE. Whenever possible, blood cultures should be obtained before starting antibiotics.
· Transthoracic echocardiography (TTE) must be supplemented with transesophageal echocar- diography (TEE) in most cases of suspected PVE. TEE is more sensitive than TTE and remains the present gold standard diagnostic modality for documenting IE. The role and added value of cardiac computed tomography (CT), magnetic resonance imaging (MRI), and other comple- mentary imaging technologies are still unclear.
· Duke criteria or modified Duke criteria are used to confirm the certainty of the diagnosis. However, clinical judgment is very important on an individual basis, such as PVE and negative blood cultures, and so on (Table 15.1).7
· Surgical treatment should be considered for patients with signs of heart failure, severe valve dysfunction, PVE, invasion with paravalvular abscess or cardiac fistulas, recurrent systemic embolization, large mobile vegetations, and persistent sepsis despite adequate antibiotic therapy for more than 5 to 7 days. Most patients with PVE will require surgery. See the next section for indications for surgery.
· Early surgery is recommended. Once a surgical indication is present, surgery should not be delayed. Early surgery is defined as being carried out “during initial hospitalization independently of completion of a full therapeutic course of antibiotics.”1,3,5
· All patients with IE who require surgery but have neurologic symptoms should have a neurologic evaluation and brain imaging by CT or MRI before the planned operation. Imaging may need to be repeated in case of new or worsening symptoms.
· In general, surgery should be delayed for 1 to 2 weeks for patients with nonhemorrhagic strokes and 3 to 4 weeks for patients with hemorrhagic strokes. For those with nonhemorrhagic embolic strokes, earlier intervention may be justified. The risk of the worsening of stroke symptoms must be weighed against the indications for surgery and risk of additional emboli during the waiting period, in consultation with a neurologist.1-5,16,19-31
· The need for preoperative coronary angiography should be guided by normal criteria. CT angiography is an alternative to assess coronary anatomy in patients with large aortic valve vegetations.1-5
· Aortic or mitral IE or PVE with severe acute regurgitation or valve obstruction causing refractory pulmonary edema or cardiogenic shock.
· Aortic or mitral IE with severe acute regurgitation or valve obstruction and persisting heart failure or echocardiographic signs of poor hemodynamic tolerance (early mitral closure of pulmonary hypertension).
· Aortic or mitral IE or severe prosthetic dehiscence with severe regurgitation and no heart failure.
· Locally uncontrolled infection (e.g., abscess, pseudoaneurysm, fistula, enlarging vegetation).
· Persisting fever and positive blood cultures more than 7 to 10 days not related to an extracardiac cause.
· Infection caused by fungi or multiresistant organisms.
· PVE caused by staphylococci or gram-negative bacteria (most cases of early PVE).
· Aortic or mitral IE or PVE with large vegetations (> 10 mm) following one or more embolic episodes despite appropriate antibiotic therapy.
· Aortic or mitral IE or PVE with large vegetations (> 10 mm) and other predictors of complicated course (e.g., heart failure, persistent infection, abscess).
· Aortic or mitral or PVE with isolated very large vegetations (> 15 mm).
· Objectives of IE surgery are to prevent additional embolic events, débride and remove all infected and necrotic tissue and foreign material, and restore functional valve and cardiac integrity.
· Adequate surgical débridement requires good surgical exposure. A median sternotomy is required for most IE operations. Ministernotomy and right thoracotomy approaches are likely to provide insufficient exposure if unexpected or more advanced and invasive disease is encountered; these procedures are not recommended for IE surgery.
· Chest CT is recommended to assess the risk of sternal reentry in patients with previous cardiac surgery.32 When an arterial structure such as an ascending aorta, pseudoaneurysm, or important graft is in direct contact with the sternum, consideration should be given to peripheral cannulation and the institution of cardiopulmonary bypass before sternotomy.5,32,34 Intraoperative TEE is mandatory.
· Perfect myocardial protection is critical because the procedure is often long and complex. This is achieved with initial induction with antegrade and retrograde blood cardioplegia and repeat retrograde cardioplegia every 15 to 20 minutes. Open insertion of the retrograde coronary sinus cannula secures perfect cardioplegia delivery.
· For those infections limited to the native valve cusps or leaflets (so-called simple IE), valve repair or replacement with a biologic or mechanical valve prosthesis according to similar principles as for patients with noninfected valves should be done. For very sick patients and those with neurologic complications, a biologic valve is recommended to avoid added anticoagulation-related complexity in postoperative management.
· For infection beyond the cusp or leaflets (advanced pathologies), radical débridement and reconstruction may be required. Radical débridement means complete removal of foreign material, necrotic tissue, and vegetations; it does not mean excision with wide margins, which may cause additional damage, jeopardize valve repair, and make reconstruction more difficult.
· All infected areas must be opened, unroofed, and cleaned out. In patients with PVE, débride- ment should include removal of the old prosthesis and suture material.
· Dirty noncardiotomy suction should be used for the initial débridement and irrigation. The use of cardiotomy suction is avoided when the field is grossly contaminated to minimize blood contamination. Débridement is followed by generous irrigation. Surgical instruments and gloves should be exchanged after the completion of débridement and irrigation.
· The excised valve specimens should be handled properly and divided between pathology and microbiology. Molecular testing of the excised cardiac valves with a polymerase chain reaction (PCR) assay should be considered when there is uncertainty regarding the causative microorganism.
· For limited localized infection and preserved cusp contour, repair may occasionally be possible. The resulting cusp defect after the removal of vegetation is repaired with an autologous pericardial patch. In most cases, cusp preservation is not possible, and valve replacement is required. The choice is based on the usual criteria, as mentioned earlier.
· For infection limited to the native aortic valve cusps, complete removal of the native valve cusps and replacement with a valve prosthesis should suffice. Extraaortic invasion of native valve endocarditis is usually localized, and subcommisural invasion is most common. Often, the site of annulus penetration is small, which hides a widely spread extraaortic infection that requires unroofing of the entire infected area for adequate débridement.6
· For invasive pathology infection beyond the valve cusps involving the annulus radical resection of all infected tissue and foreign material is necessary. Adequate débridement is followed by reconstruction (Fig. 15.1).
· We recommend caution so as not to lose track of anatomy, cause injury to coronary arteries, or sacrifice live left ventricular outflow tract (LVOT) muscle and surrounding structures to make reconstruction more difficult and risky.
· When additional material is required for reconstruction, autologous pericardium is our preference, but bovine pericardium or other materials can be used. Even in patients with invasive disease, the tissue destruction usually leaves the LVOT intact, and no additional material is required for the reconstruction in most cases.6,9
· For invasive disease requiring aortic root reconstruction, an aortic allograft is our preferred choice. The more extensive and destructive the infection, the stronger is the argument in favor of an allograft over alternative conduits with prosthetic valves.5,12
· Bioroots (bioprosthetic valve inside a graft), mechanical valve conduits, porcine aortic roots, and bovine pericardial root reconstruction may also work if the allografts are unavailable. This is also true for aortic PVE. The use of an allograft is no substitute for the radical débridement of all infected tissue!
· See Fig. 15.1.a
· A prosthetic aortic valve usually involves the sewing ring and, in contrast to native valve endocarditis, the invasion is often circumferential. Although the deeper invasion and tissue destruction can be anywhere around the annulus circumference, large root abscesses develop preferentially posteriorly and to the left, under the pulmonary trunk.
· Bacterial invasion from the aortic root works its way from posterior aortic root invasion into the right atrium and triangle of Koch; destroying the atrioventricular node and upper end of the bundle of His is the most common cause of heart block in IE. If a patient has heart block of any degree, the right atrium must be opened for inspection.6
· Occasionally. the sewing ring is infected, but the infection has not yet penetrated deeper into the annulus. In these cases, it is sometimes feasible to perform adequate débridement and implant another prosthetic valve of choice without the need for root replacement.
· More commonly, the infection in PVE needs more extensive débridement and root reconstruction. This is done in a similar fashion as that discussed earlier, in the aortic NVE section (see Fig. 15.1).
Figure 15.1Prosthetic valve endocarditis with sepsis and heart block. (A) Infected mechanical prosthesis with vegetations on sewing ring (arrow). (B) Same patient with perforation visible in right atrium (RA; arrow). (C) After debridement, destruction in location of atrioventricular node is seen. This infection has worked its way around the aorta counterclockwise over an extended period, displaying a pseudoaneurysm stage anteriorly and an active cellulitis stage posteriorly and into right atrium. Left ventricular outflow tract (LVOT) is intact and ready for reconstruction. (D) After complete debridement of all infected tissue, RA is reconstructed with autologous pericardium (arrow). (E) Aortic allograft is sutured to LVOT with running monofilament suture. (F) Allograft is tied down and well seated, allowing debrided infected areas to communicate and drain to pericardium. CFB, central fibrous body; CS, coronary sinus; LCA, left coronary artery; RCA, right coronary artery; TV, tricuspid valve.
· The LVOT is almost always preserved after extensive débridement to allow direct anastomosis to the allograft.6
· The main landmarks for guiding the reconstruction and indicating the level of the proximal suture line are the intervalvular fibrosa (IVF) corresponding to the base of the anterior mitral leaflets and the two trigones on either side. Both coronary buttons should be adequately mobilized and large enough for any future reoperation.
· The LVOT is sized with Hegar dilators, and an allograft with an internal diameter 2 to 3 mm less than the diameter of the annulus is chosen. Correct sizing is important.
· If a smaller allograft is unavailable, the annulus size is reduced by placing two 2-0 Gore-Tex sutures around the annulus and tying them down over a Hegar dilator. We avoid the use of felt or additional support material for the suture line. The allograft is implanted in an anatomic orientation.
· The proximal suture line (between the allograft and LVOT) is performed with running 3-0 or 4-0 monofilament sutures, allowing seating of the allograft deep inside the annulus. A running technique instead of an interrupted technique allows the distribution of tension equally to all suture loops.
· The allograft is lowered into the LVOT with gentle traction on the sutures, and perfect seating is ensured. The coronary buttons are reimplanted on the allograft in anatomic positions using running 4-0 or 5-0 monofilament sutures.
· The distal anastomosis (allograft to aorta) is performed with running 4-0 monofilament sutures. The length of the allograft should be generous to allow for tension-free anastomosis on either end.
· Reconstruction of a destroyed IVF in advanced aortic root destruction is discussed separately.
· Mitral valve endocarditis has some specific features related to its anatomy and degenerative pathologic features. This makes radical débridement more difficult to accomplish in mitral cases with atrioventricular groove invasion, necrosis, and abscess formation. This often means sealing off the infected and débrided cavity, with a resulting increased risk of recurrent infection.
· Mitral annular calcium is frequently the starting site of both infection and invasion. Invasive disease is less common with mitral than with aortic valve endocarditis and, when invasion occurs, it is often shallow. Invasion of the anterior annulus leads to destruction of the subaortic curtain; invasion into the posterior annulus leads to entry into the atrioventricular groove and separation of the atrium from the ventricle.
· The mitral valve is exposed via a left atriotomy through the interatrial groove (Sondergaard’s groove) or transseptally through the right atrium, which we prefer. If the left atrium is small, an extended transseptal dome approach can be used for increased exposure.
· Dual exposure via an aortotomy is helpful in some cases for débridement and suture placement and to avoid aortic valve injury.
· All grossly infected tissue is removed and the unaffected leaflet, chordae, and papillary muscles are preserved to support the posterior annulus. Mitral valve repair is preferred and can be performed safely as long as sufficient tissue remains to allow reconstruction (Figs. 15.2 and 15.3). Standard mitral valve repair techniques are used.
· A prosthetic mitral annuloplasty ring or band has a very low added risk of recurrent infection and can be safely used to provide durable repair. If repair is not possible, the valve needs to be replaced. The choice of prosthesis follows the normal principles of valve surgery.
· In case of invasive disease requiring reconstruction of the mitral annulus, the patches (usually autologous or bovine pericardium) must be generous to minimize stress on the suture lines. Relatively small lesions on the anterior leaflet (so-called kissing lesions, typically in association with aortic valve IE) require débridement and repair with autologous pericardium using running polypropylene sutures.
· In patients with extensive disease involving destruction of the aortic valve and mitral valve along the base of the anterior mitral leaflet, an aortic allograft with an attached anterior leaflet of mitral valve provides additional benefit. It can be used to repair the defect in the anterior mitral leaflet and reconstruct the aortomitral curtain.
· Localized defects after débridement of the posterior leaflet can be treated by triangular or quadrangular resection. A sliding repair can be added, if required.
Figure 15.2 Mitral valve endocarditis with large vegetation on posterior mitral valve leaflet, with leaflet perforation. Patient had a preoperative embolic stroke. Valve was repaired after excising the vegetation.
Figure 15.3 (A) Mitral valve endocarditis involving the medial trigone. (B) After resection and repair with a pericardial patch and an anuloplasty ring.
· Unlike prosthetic aortic valve endocarditis, the exposure for débridement and removal of the old prosthesis and suture material is worse for prosthetic mitral valve endocarditis. A dual approach via the left atrium and aorta, as described earlier, is very helpful.
· Use a generous patch (to minimize tension on the suture line) if annulus reconstruction is required. Anchorage to the ventricular muscle to prevent communication and entry into the paravalvular cavities beneath the valve is very important (Fig. 15.4).
· David’s technique uses a semicircular pericardial patch for annular reconstruction, with one side of the patch secured to the endocardium of the left ventricle and the other side secured to the left atrium. The new prosthesis is then affixed to the reconstructed annulus.
· In case the atrioventricular separation is shallow and narrow, the suture closure technique can be used.46 Valve sutures are then placed with pledgets on the ventricular side.
· Most cases of endocarditis involving both the aortic and mitral valves can be managed in a manner similar to what has been discussed for each of these valves separately. Destruction of the IVF or aortomitral curtain requires reconstruction, which is technically demanding and is a high-risk surgery (Fig. 15.5). IVF destruction usually occurs in the setting of PVE affecting both the aortic and mitral valves, but can occur in any combination or can be extensive disease of the aortic or mitral valve, with extension into the IVF.
· Excellent exposure is required. This can be accomplished by using an extended transseptal approach or by dividing the superior vena cava and extending the left atriotomy toward the dome of the left atrium. This therefore allows excellent exposure for débridement of the aortic and mitral valves, as well as the IVF. Often, however, an incision in the dome and IVF is enough.
· Débridement is followed by generous irrigation of the operative field. The mitral prosthesis is sized. The IVF corresponds to one-third of the circumference, and the posterior annulus from trigone to trigone corresponds to two-thirds of the circumference.
· The mitral valve prosthesis is implanted first. Valve sutures are placed posteriorly from trigone to trigone, with pledgets on the ventricular side. Two-thirds of the mitral valve prosthesis is secured posteriorly at this time.
· The IVF is reconstructed using any available tissue or patch material autologous or bovine pericardium, synthetic material or by direct implantation of an aortic allograft by suturing the allograft mitral valve directly to the mitral valve prosthesis. If using a patch material, a generous double-layered patch is sewn to the mitral prosthesis’ sewing ring and anchored to both the trigones. It is critical to secure tension-free closure of the corner where the mitral annulus, left ventricular wall, and aortic annulus meet.
· The lower sheet of the patch is used to close the dome of the left atrium, and the upper sheet is used for reconstruction of the aortic root. The aortic prosthesis is implanted by securing it to the native aortic annulus and to the patch.
· As discussed earlier, when an aortic allograft is used, the allograft mitral valve is directly sewn to the mitral prosthesis. A separate patch to close the left atrial dome may still be required. The most critical area for bleeding is from the lateral trigone and requires tension-free reconstruction.
· Postoperative complications in patients undergoing surgery for active endocarditis are common. Postoperative sepsis is very common in these patients. They commonly present with vasoplegia and hypotension. A combination of sepsis, prolonged cardiopulmonary bypass times, and extensive débridement and major reconstruction may cause severe coagulopathy and excessive bleeding in the postoperative period.
· Our recommended way of dealing with postoperative coagulopathy in the operating room is as follows: controlling any surgical bleeders before giving protamine, packing and avoiding suctioning for 20 to 30 minutes after protamine to allow for clotting before attempting additional surgical hemostasis, and use of blood products as required.
· All patients with active endocarditis receive postoperative antibiotics; the standard duration is 6 weeks when the infection is active at the time of surgery. The duration of therapy may be modified by specific clinical scenarios or organisms; thus, antibiotic treatment and its duration should be carried out in consultation with an infectious disease specialist. In patients with fungal endocarditis, we recommend a lifelong oral antifungal for suppression because we have seen numerous recurrences after the antifungal has been stopped.
· Postoperatively, patients should be reviewed for probable sources of bacteremia, depending on the specific causative microorganism. Apart from teeth and mouth, patients with Streptococcus gallolyticus often have colon polyps or colon cancer and should undergo a colonoscopy.
· All patients should have an echocardiogram before discharge to verify the surgical repair and establish a baseline echocardiogram for follow-up.
· Despite significant improvement in surgical results, in-hospital mortality for a patient undergoing IE surgery remains higher than for any other valve surgery. Even with appropriate antibiotics and surgical intervention, reported in-hospital mortality is 15–20%, and 1-year mortality has approached 40%.1-5 Traditional factors predicting worse outcomes in endocarditis surgery have been PVE, invasive stage, which includes abscesses, and S. aureus.
· High-volume centers with extensive experience treating endocarditis have more recently reported lower hospital mortality for surgically treated left-sided endocarditis (8% 30-day mortality). These centers have also demonstrated similar improved survival for NVE as well as PVE.5,9,12
· Surgically treated left-sided invasive IE has a worse hospital mortality rate than noninvasive cases (11% vs. 4.4%), mainly because of invasive mitral valve disease. The outcomes after surgery for aortic valve IE were similar whether or not the disease was invasive (Fig. 15.6).9
· For mitral valve IE, event-free survival, hospital mortality, and long-term survival are all superior after mitral valve repair compared to replacement.38-45
· The objectives of endocarditis surgery are to débride and remove all infected and necrotic tissue and foreign material and restore functional valves and cardiac integrity.
· Other key surgical principles are generous irrigation after débridement, the use of an allograft for invasive aortic valve endocarditis, and avoiding the use of additional foreign material and an adhesive (e.g., BioGlue).
· In the end, however, the choice of prosthesis is less crucial for the surgical outcome than the need for complete débridement.
· The optimal management of patients with an embolic stroke or cerebral hemorrhage remains a difficult challenge with regard to optimal timing of surgery.
· Surgery for endocarditis is indicated as soon as a surgical indication is present. Earlier surgery prevents an additional embolism and avoids further destruction and invasion.
1. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2015;132:1435–1486.
2. Habib G, Lancellotti P, Antunes MJ, et al. 2015 ESC guidelines for the management of infective endocarditis: The Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur Heart J. 2015;36:3075–3128.
3. Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129:2440–2492.
4. Byrne JG, Rezai K, Sanchez JA, et al. Surgical management of endocarditis: the Society of Thoracic Surgeons clinical practice guideline. Ann Thorac Surg. 2011;91:2012–2019.
5. Pettersson GB, Coselli JS, Hussain JT, et al. 2016 The American Association for Thoracic Surgery (AATS) consensus guidelines: surgical treatment of infective endocarditis: executive summary. J Thorac Cardiovasc Surg. 2017;153:1241–1258.
6. Pettersson GB, Hussain ST, Shrestha NK, et al. Infective endocarditis: an atlas of disease progression for describing, staging, coding and understanding the pathology. J Thorac Cardiovasc Surg. 2014;147:1142–1149.
7. Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30:633–638.
8. Cahill TJ, Baddour LM, Habib G, et al. Challenges in infective endocarditis. J Am Coll Cardiol. 2017;69:325–344.
9. Hussain ST, Shrestha NK, Gordon SM, et al. Residual patient, anatomic, and surgical obstacles in treating active left-sided infective endocarditis. J Thorac Cardiovasc Surg. 2014;148:981–988.
10. Gaca JG, Sheng S, Daneshmand MA, et al. Outcomes for endocarditis surgery in North America: a simplified risk scoring system. J Thorac Cardiovasc Surg. 2011;141:98–106, e1-2.
11. David TE, Gavra G, Feindel CM, et al. Surgical treatment of active infective endocarditis: a continued challenge. J Thorac Cardiovasc Surg. 2007;133:144–149.
12. Manne MB, Shrestha NK, Lytle BW, et al. Outcomes after surgical treatment of native and prosthetic valve infective endocarditis. Ann Thorac Surg. 2012;93:489–493.
13. Thuny F, Grisoli D, Collart F, et al. Management of infective endocarditis: challenges and perspectives. Lancet. 2012;379:965–975.
14. Prendergast BD, Tornos P. Surgery for infective endocarditis: who and when? Circulation. 2010;121:1141–1152.
15. Thuny F, Beurtheret S, Mancini J, et al. The timing of surgery influences mortality and morbidity in adults with severe complicated infective endocarditis: a propensity analysis. Eur Heart J. 2011;32:2027–2033.
16. Kang DH, Kim YJ, Kim SH, et al. Early surgery versus conventional treatment for infective endocarditis. N Engl J Med. 2012;366:2466–2473.
17. Savage EB, Saha-Chaudhuri P, Asher CR, et al. Outcomes and prosthesis choice for active aortic valve infective endocarditis: analysis of the Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2014;98:806–814.
18. Bedeir K, Reardon M, Ramlawi B. Infective endocarditis: perioperative management and surgical principles. J Thorac Cardiovasc Surg. 2014;147:1133–1141.
19. Funakoshi S, Kaji S, Yamamuro A, et al. Impact of early surgery in the active phase on long-term outcomes in left-sided native valve infective endocarditis. J Thorac Cardiovasc Surg. 2011;142:836–842, e1.
20. Chu VH, Park LP, Athan E, Delahaye F, et al. Association between surgical indications, operative risk, and clinical outcome in infective endocarditis: a prospective study from the International Collaboration on Endocarditis. Circulation. 2015;131:131–140.
21. Hodges KE, Hussain ST, Stewart WJ, Pettersson GB. Surgical management of infective endocarditis complicated by ischemic stroke. J Card Surg. 2017;32:9–13.
22. Yanagawa B, Pettersson GB, Habib G, et al. Surgical management of infective endocarditis complicated by embolic stroke: practical recommendations for clinicians. Circulation. 2016;134:1280–1292.
23. Hess A, Klein I, Iung B, et al. Brain MRI findings in neurologically asymptomatic patients with infective endocarditis. AJNR Am J Neuroradiol. 2013;34:1579–1584.
24. Duval X, Iung B, Klein I, et al. Effect of early cerebral magnetic resonance imaging on clinical decisions in infective endocarditis: a prospective study. Ann Intern Med. 2010;152:497–504.
25. Barsic B, Dickerman S, Krajinovic V, et al. Influence of the timing of cardiac surgery on the outcome of patients with infective endocarditis and stroke. Clin Infect Dis. 2013;56:209–217.