| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2007;6:354-357. doi:10.1510/icvts.2007.151621 © 2007 European Association of Cardio-Thoracic Surgery
Early surgery for hospital-acquired and community-acquired active infective endocarditisDepartment of Cardiovascular Surgery, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno, Osaka, 545-8585, Japan Received 4 January 2007; received in revised form 4 March 2007; accepted 6 March 2007
*Corresponding author. Tel.: +81-6645-3980; fax: +81-6646-3071.
Active infective endocarditis (IE) is classified into two groups; hospital acquired IE (HIE) and IE other than HIE, which was defined as community-acquired IE (CIE). Eighty-two patients underwent surgical treatment for active IE. Seventy-one cases were CIE group and eleven were HIE. There were six patients with native valve endocarditis and five cases of prosthetic valve endocarditis in the HIE group. We compared the surgical outcome of both types of active IE retrospectively. The preoperative status of the patients in the HIE group was more critical than that in the CIE group. Streptococcus spp. were the major micro-organisms in the CIE group (39%), while 82% of the HIE cases were caused by Staphylococcus spp. All Staphylococcus organisms in the HIE group were methicillin resistant. There were 10 hospital deaths, three in the CIE group and seven in the HIE group. Operative mortality in the HIE group was significantly higher than in the CIE group (63.6% vs. 4.2%, P<0.001). The outcome of early operation was satisfactory for active CIE, but poor for HIE. These types of active IE should be considered separately.
Key Words: Endocarditis; Hospital-acquired infection
Surgical treatment of active infective endocarditis (IE) still has high mortality and morbidity, and the most controversial issue is when we should operate for active IE [1–4]. Recently, early operation for active IE was proposed without prolonged antibiotic therapy [5]. We have encouraged early surgery for active IE when the patient meets the Duke and Manhas criteria [6, 7]. Our surgical approach to the patient with active IE is (1) to operate before hemodynamic deterioration occurs if there is severe regurgitation; (2) to operate earlier if echocardiography reveals a vegetation (larger than 1.0 cm in diameter) that is considered to pose a high risk for embolism [8, 9]. Active IE should be classified into two groups: hospital-acquired IE (HIE) [10, 11] and IE other than HIE. We defined the latter as community-acquired IE (CIE) in this study. Newer invasive therapeutic interventions have increased the risk of bacteremia in the population at risk, which has led to an increasing risk of HIE. Although both types of IE are diagnosed with the same criteria, we should consider HIE and CIE separately because the outcomes of treatment are different [12]. In this study, retrospective reviews were performed to evaluate the results of surgical intervention for active HIE and CIE.
2.1. Patients Between April 1992 and September 2005, 82 consecutive patients underwent surgical treatment for active IE in our institute. The clinical diagnosis of IE was established in all patients using the Duke criteria [6]. The criteria for active IE followed the report of Manhas [7]. Native valve endocarditis was present in 62 patients and prosthetic valve endocarditis (PVE) in 20. Eleven patients had HIE and 71 had CIE. HIE is defined as infective endocarditis occurring 48 h or more after admission, or endocarditis that is related to an intervention performed in hospital within 4–8 weeks before admission. Early PVE is, therefore, HIE fulfilled by the latter definition. Although we recognize that early PVE is a unique clinical entity, cases of early PVE during the same hospitalization were included with HIE in this study. The HIE group included five patients with early PVE during hospitalization and six patients with native valve infection. Infection of the artificial ring in mitral annuloplasty was included with PVE in this study. Antibiotics that were shown to be sensitive by blood culture test were used before and after surgery. When blood cultures were negative, penicillin and aminoglycoside were used empirically. Antibiotic therapy was continued for at least 4 weeks after surgery. The timing of operation mainly depended on the cardiologists' decision. Patients who had collapsed hemodynamics or large flail vegetation underwent emergent or urgent operation. Continuous data were expressed as mean±standard deviation (S.D.) or median with the range. Categorical data were compared using Fisher's exact test. Continuous variables were compared with the t-test or Mann–Whitney U-test.
3.1. Preoperative clinical features The patients' characteristics and preoperative features are shown in Table 1. Patients in the HIE group were more likely to have infection of artificial prostheses, since early PVE was included in this group. The preoperative condition of patients was more critical in the HIE group, with respiratory failure and disseminated intravascular coagulopathy (DIC), than in the CIE group. The median durations of fever were eight weeks and two weeks in the CIE group and HIE group, respectively. The duration of fever in the CIE group was significantly longer than that in the HIE group. The percentage of the patients who had fever longer than four weeks was 61% in the CIE group and 16% in the HIE group.
3.2. Microbiological features The micro-organisms causing the infections are summarized in Table 2. Sixty-nine patients (84%) had culture-positive endocarditis and 13 were culture negative at the time of operation. Streptococcus spp. were the major micro-organisms in the CIE group, while 82% of micro-organisms in the HIE group were Staphylococcus spp. All Staphylococcus spp. in the HIE groups were methicillin resistant.
3.3. Position of infected and operated valves The positions of the infected valve are summarized in Table 3. Two of the four tricuspid valve infections were in drug addicts. There were 16 patients of prosthetic valve endocarditis in the CIE group and five in the HIE group. The previous valve surgery were seven AVR, four MVR, four Bentall, and one Mitral valve repair in the CIE group. The median duration from the previous valve surgery to re-do surgery for PVE was 56 weeks and four weeks in the CIE and the HIE groups, respectively. Bioprosthetic valves were more likely used in the HIE group. Mitral valve repair had been performed in four patients (three with an annuloplasty ring, one without) in the CIE group. Five patients (7%) in the CIE group had aortic root abscess, and one patient in the HIE group had an annular abscess of the mitral valve. The incidence of urgent or emergency operation was 70% in the CIE group and 55% in the HIE group.
Aortic root reconstructions were done in nine patients in the CIE group. One patient with aortic root abscess had AV block preoperatively. Four patients had an infection of the aortic root after a modified Bentall operation. Eight patients underwent aortic root replacement with a stentless porcine valve, and one underwent a modified Bentall operation with a composite graft. 3.4. Operative mortality and morbidity There were 10 hospital deaths, three in the CIE group and seven in the HIE group. Operative mortality in the HIE group was significantly higher than that in the CIE group (63.6% vs. 4.2%, P<0.001). The causes of death in the CIE group were cerebral bleeding (n=2) and low output syndrome (n=1). There were three valve-related deaths and one recurrence of infection in the CIE group during the long-term follow-up period. All five patients in the HIE group who were discharged from hospital were doing well. Table 4 shows the clinical characteristics of the HIE group. Cases 1–6 had native valve infections. Four patients were on a respirator and two had DIC before operation. Cases 7–11 had PVE including artificial ring infection.
The micro-organisms demonstrated by blood cultures were methicillin-resistant Staphylococcus or fungus. In Case 4, the blood cultures were negative, but vegetations were detected before the operation by transesophageal echocardiography (TEE). The duration from previous intervention therapy and cardiac surgery for IE was 5.5 weeks on average. Four patients who underwent emergency surgery for unstable hemodynamics died. Patients who were in multiple organ failure also died. There was one death due to brain complications (Table 5).
This study revealed that the surgical outcome of HIE differs from that of CIE. Mortality in HIE is quite high, though that in CIE is acceptable. The mortality in the CIE group was 4% in this study, which was a little higher than that of the ordinal valve surgery. Radical debridement of infected tissues and restoration of valve function are fundamental tenets of surgical treatment of IE. Proper choice of antibiotics is also important. These may lead to satisfactory surgical results. We have encouraged early operation for active IE, and the results of this study show the validity of our strategy for active IE in patients with CIE. In this study, the major differences between HIE and CIE were the preoperative condition of the patients and the micro-organisms involved. Bacteremia sometimes occurs in severely ill patients. Usually, it takes a long time for physicians to recognize the presence of IE in patients who are treated in hospital due to non-cardiac disease, which causes severe illness in most HIE patients. As shown in Table 1, the preoperative condition of HIE patients was worse than that of CIE patients. When a compromised patient is in sepsis, the physician should manage the patient considering the possibility of IE at any time. The difference between the micro-organisms is the second important issue. The major micro-organisms in HIE are Gram-positive cocci [13]. In the HIE group, all Staphylococcus spp. were methicillin resistant. Especially in methicillin-resistant Staphylococcus aureus infection, the valves have already been destroyed and the general condition of patients has been severely impaired before IE is recognized. Blood cultures are necessary to identify the micro-organisms before antibiotic therapy; however, our results might allow us to use vancomycin or teicoplanin empirically when we suspect HIE. PVE has high mortality [2, 4]. The clinical course of early-onset PVE is typically more aggressive and fulminant, with high mortality. Castillo et al. reported that the mortality of early PVE in hospital was 31%, whereas it was 9% in late PVE [4]. We had four cases of infection of an annuloplasty ring after mitral valve repair in this study. An annuloplasty ring is now used in most cases of mitral repair. Endocarditis affecting the repaired mitral valve has similarities with both native valve endocarditis and PVE. Gordon and colleagues reported that the incidence of early-onset endocarditis of the annuloplasty ring was 0.2% [5]. Gillinov and colleagues mentioned that early re-operation may be unnecessary, because antibiotics alone can eradicate infection that is limited to a leaflet [14]. However, it is difficult to prove whether the annuloplasty ring is infected on transthoracic echocardiography. Indeed, though one patient in the present study (Case 8) was blood culture positive, frequent echocardiography failed to detect vegetations on the mitral annuoplasty ring, which led to prolonged antibiotic therapy before re-operation. In our patients, urgent surgery was necessary due to vegetations on the annuloplasty ring, which were demonstrated by TEE. We believe in the usefulness of frequent TEE when IE is suspected. The choice of valve substitute is another issue in the management of active IE. In this study, the choice of biological or mechanical valve was made according to the patient's age or compliance with anticoagulation therapy after operation. Most patients in the HIE group were so severely impaired that they could not take warfarin early after surgery. This was the major reason why more bioprostheses were used in the HIE patients. It is difficult to obtain homografts in Japan, so we cannot help but use stentless valves instead. We have reported previously that stentless valves are useful for the treatment of destroyed aortic root [15]. The outcome of early operation for active CIE was satisfactory in this study. In contrast, the outcome of surgery in patients with HIE was poor, especially in those with multiple organ failure or DIC preoperatively. Our results encourage early surgery for CIE. In order to improve the surgical outcome of HIE, we would need earlier detection of vegetation by echocardiography before deterioration of general condition in HIE patients. In conclusion, we should recognize that the characteristics and surgical outcome of CIE and HIE are quite different.
The authors wish to thank Dr. Mitsuru Fukui (Osaka City University) for his contributions to the statistical assistance.
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
