Infective Endocarditis Diagnosis, Antimicrobial Therapy, and ...

Prediction of individual patient risk for embolization isextremely difficult. Many studies have attempted to useechocardiography to identify a high-risk subset of IE patientswho might benefit from early surgery to avoid embolization.Several studies with TTE have demonstrated a trend towardhigher embolic rates with left-sided vegetations 1 cmindiameter. 50 DeCastro and colleagues 225 compared TTE withmultiplane TEE and found that neither technique was helpfulin defining embolic risk in patients with vegetations. In astudy 52 based on TEE, mitral vegetations 1 cm in diameterwere associated with the greatest frequency of embolism. Theassociation was strengthened when analysis was limited tothose patients who had not yet experienced a clinical embolicevent. Another prospective TEE study, however, found noclear correlation of vegetation size with embolization. 54Overall, these data are compatible with previous observationsthat indicate that in general, mitral vegetations of any size areassociated with a higher risk of embolization (25%) thanaortic vegetations (10%). As noted above, the highest embolicrisk (37%) has been seen in the subset of patients withmitral vegetations attached to the anterior rather than theposterior mitral leaflet. 54,58 This suggests that the mechanicaleffects of broad and abrupt leaflet excursion, occurring twiceper heartbeat, may contribute to the propensity of a vegetationto fragment and embolize.In another study, the effect of vegetation size on embolicpotential was dependent on the infecting organism, with largevegetations independently predicting embolic events only inthe setting of streptococcal IE. 54 In contrast, as confirmedabove by Vilacosta et al, 227 staphylococcal or fungal IEappears to carry a high incidence rate of embolizationindependent of vegetation size.The role of echocardiography in predicting embolic eventshas been controversial. In 1 survey 228 that included 4 echocardiographerswho were blinded to clinical data, interobserveragreement was mixed on the characterization ofvegetations. Agreement was high for the presence of vegetation(98%) and involved site (97%); interobserver agreementwas considerably less for vegetation size (73%), mobility(57%), shape (37%), and attachment (40%).An increase in vegetation size over 4 to 8 weeks of therapyas documented by TEE does appear to predict embolic events.In addition, a second, albeit infrequent, peak of late embolicevents has been observed to occur 15 to 30 weeks after thediagnosis of IE and has been associated with nonhealingvegetations (failure of a vegetation to stabilize or diminish insize) as defined by echocardiography. 58The traditional indications for valvular surgery for IE toavoid embolization have been 2 major embolic events. 229These criteria are arbitrary and exclude cutaneous embolization,which is common, or embolism occurring before theinstitution of therapy. 229 Because of the observed decreases inembolic risk during the first 2 weeks of antibiotic therapy, thebenefit of surgery in avoiding catastrophic embolic events isgreatest early in the treatment course of IE. Early surgicalintervention may preclude a primary or recurrent majorembolic event but exposes the patient to both the immediateand lifelong risks of valve replacement. At this time, thestrategy for surgical intervention to avoid systemic embolie420Circulation June 14, 2005The decision to operate on patients with IE is drivenprimarily by severity of CHF. Poor surgical outcome ispredicted by preoperative New York Heart Association ClassIII or IV CHF, renal insufficiency, and advanced age. In anypatient, a decision to delay surgery to extend preoperativeantibiotic treatment carries the risk of permanent ventriculardysfunction and should be discouraged. The incidence ofreinfection of newly implanted valves in patients with activeIE has been estimated to be 2% to 3%, 218,219 far less than themortality rate for uncontrolled CHF.Surgical approaches to IE patients with CHF must take intoaccount the distortion of the valve and its surroundingstructures. Severe valvular disruption usually will requireprosthetic replacement. Ruptured mitral chordae may sometimesbe repaired with a combination of leaflet resection,chordal reattachment or transposition, and annular support.Leaflet perforations may be repaired with small pericardialpatches if the surrounding leaflet tissue is well preserved andvalve motion can be maintained. Similarly, in selected cases,discrete vegetations on aortic or mitral leaflets have beenexcised along with underlying leaflet tissue (“vegetectomy”)and repaired with a patch. To date, clinical experience withvegetation excision has been predominantly limited to mitralvalve IE. Recent experiences in Europe have emphasized thepotential for early valve repairs when feasible, especially inpatients with anterior mitral valve IE and even in the absence ofother conventional indicators for surgical interventions. 220–222This approach has a 2-fold potential benefit: Because 50% ofpatients with left-sided native valve IE will require valvereplacement surgery within 10 to 15 years of the IE episode, 223this “preemptive” repair strategy has the advantage of earlier-ageintervention; and this approach may circumvent ultimate valvereplacement requirements, with the attendant risks of long-termanticoagulation.Risk of EmbolizationSystemic embolization occurs in 22% to 50% of cases ofIE. 50,52,224–226 Emboli often involve major arterial beds, includingthe lungs, coronary arteries, spleen, bowel, andextremities. Up to 65% of embolic events involve the centralnervous system, and 90% of central nervous system embolilodge in the distribution of the middle cerebral artery. 226 Thehighest incidence of embolic complications is seen withaortic and mitral valve infections and in IE caused by Saureus, Candida, HACEK, and Abiotrophia organisms. Embolican occur before diagnosis, during therapy, or aftertherapy is completed, although most emboli occur within thefirst 2 to 4 weeks of antimicrobial therapy. 227 Of note, 2independent studies have confirmed that the rate of embolicevents drops dramatically during or after the first 2 to 3 weeksof successful antibiotic therapy. In a study from 1991, theembolic rate dropped from 13 to 1.2 embolic events per1000 patient-days during that time. 52 In a more recent study,Vilacosta et al 227 confirmed the reduced frequency of embolizationafter 2 weeks of therapy. Moreover, the latter studyreemphasized the increased risk of embolization with increasingvegetation size during therapy, mitral valve involvement,and staphylococcal causes.Downloaded from circ.ahajournals.org by on June 26, 2007