HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John G. Byrne Lawrence H. Cohn Sary F. Aranki Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mihaljevic, T. Articles by Aranki, S. F. Search for Related Content PubMed PubMed Citation Articles by Mihaljevic, T. Articles by Aranki, S. F. Related Collections Valve disease

Eur J Cardiothorac Surg 2001;20:842-846
© 2001 Elsevier Science NL

Long-term results of multivalve surgery for infective multivalve endocarditis

Division of Cardiac Surgery, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, USA

Received 14 March 2001; received in revised form 5 June 2001; accepted 11 June 2001.

Tel.: +1-617-732-7678; fax: +1-617-732-6559
e-mail: sfaranki{at}bics.bwh.harvard.edu

	Abstract

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 
Objective: The natural history of medically treated multivalvular endocarditis is associated with dismal short and long term survival. However, the impact of surgical intervention on these results is relatively unknown. The objective of this retrospective study was to report our long-term results in patients requiring multivalve surgery for multivalvular endocarditis. Methods and results: Over a 24 year period beginning in 1972, multivalve surgical procedures were performed on 63 patients for infective endocarditis. Prosthetic valve endocarditis was present in 25 (40%), and acute or active endocarditis in 38 (60%). The early mortality was 16%. Out of 53 patients discharged from the hospital 87±4% were alive at 5 years and 64±9% at 10 years. There was no difference in early or late mortality between patients with prosthetic and native endocarditis (P=0.15 and P=0.77 for early and late mortality, respectively). The presence of active endocarditis did not affect operative outcome or late mortality. Twenty-one patients (88%) were in NYHA FC I, and none were in NYHA FC IV. The only prognostic factor of early and late mortality was the presence of an abscess at the time of the surgery. Conclusions: These results indicate that multivalve infective endocarditis treated surgically is associated with acceptable early and late mortality and excellent postoperative functional status. The early surgical intervention prior to an abscess formation offers the best chance for survival of patients with multivalve endocarditis.

Key Words: Endocarditis • Valves • Surgery

	1. Introduction

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 
Infective endocarditis is a serious and frequently fatal condition with an estimated 10,000 to 20,000 new cases each year in the United States [1]. Antimicrobial therapy is a mainstay of medical management that resulted in a dramatic decline in mortality and achieved a high cure rate depending on the causative microorganism [2]. The evolution of cardiovascular surgical techniques and introduction of prosthetic valves further improved the outcome of the patients refractory to the standard antibiotic therapy. The wide use of prosthetic valves resulted in a new form of endocarditis, ‘prosthetic endocarditis’ (PVE), with higher morbidity and mortality than the ‘native valve endocarditis’ (NVE). The overall incidence of endocarditis with prosthetic valves is 1–4%, which accounts for 5–15% of all cases of endocarditis [1–4].

Most articles on the surgical therapy of endocarditis were based on the results of studies of isolated native or prosthetic valve. Multivalve endocarditis is recognized as a separate risk factor associated with a poor final outcome of surgical therapy. However, despite its clinical importance, it has not been studied as a separate clinical entity.

The purpose of this retrospective study was to summarize and present our experience with multivalve surgery for infective endocarditis and to determine the factors affecting the outcome of the surgical therapy.

	2. Material and methods

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 
2.1. Patient population
All patients undergoing multivalve replacement for infective endocarditis at the Brigham and Women's Hospital between January 1972 and May 1996 were the subjects of this study; prior cardiac surgery did not exclude patients’ entry. During a 24-year period between January 1972 and May 1996, 63 patients underwent multivalve surgery for infective endocarditis. Patient age ranged from 18–75 years (mean age 49 years); 24 patients (38%) were women. A large proportion of the patients was in New York Heart Association functional classes IV (31 patients (49%)).

Bioprostheses implanted during the study period included Carpentier–Edwards and Hancock valves, mechanical prostheses included Starr–Edwards, Bjork–Shiley, and St. Jude valves.

2.2. Definitions
The diagnosis of endocarditis was based on defined clinical criteria including appropriate combination of positive blood cultures, clinical signs, and/or histologic confirmation of endocarditis at operation or autopsy. Echocardiographic imaging of vegetations, valve dysfunction, or perivalvular abscesses provided further diagnostic evidence for the presence of endocarditis [5].

Endocarditis was labeled as ‘active’ if the patient required operation prior to the completion of the standard course of antibiotic therapy. The duration of therapy was variable and dependent on the causative microorganism and severity of sepsis, averaging 6–8 weeks. The term ‘healed’ endocarditis was used if the surgery was performed after the completion of antibiotic treatment. ‘Culture-negative’ endocarditis was defined by the absence of positive culture in patients presenting with clinical signs of endocarditis. The diagnosis was confirmed at operation by the presence of surgical findings of endocarditis i.e. presence of vegetation, leaflet perforation, valvular or perivalvular tissue destruction. All resected surgical specimens were cultured.

Prosthetic valve endocarditis was defined as an infection occurring on any type of mechanical or tissue valve or a reconstructed native valve [6]. By convention, PVE is termed ‘early’ when it appears within 60 days of valve insertion. Endocarditis occurring after 60 days was termed ‘late’.

2.3. Operations
Indications for surgery included refractory heart failure related to valvular dysfunction, persistent bacteriemia or fungal infection despite intensive medical therapy, myocardial or perivalvular abscess, peripheral systemic emboli, large vegetations seen on echocardiography and perivalvular leak for PVE.

All operations were performed with the use of systemic hypothermic cardiopulmonary bypass and moderate hemodilution. Blood cardioplegia with both antegrade and retrograde delivery was increasingly used since 1990.

Patients with infection limited to the leaflets of native or bioprosthetic valve were treated by valve replacement except for cases of mitral or tricuspid valve endocarditis where reconstruction was feasible. If infection had extended into or beyond annulus, aggressive debridement of all infected and inflamed tissue and reconstruction of the left ventricular inflow and outflow tracts. In cases of destruction of the aortic annulus aortic homograft was implanted.

2.4. Data collection and statistical analysis
Data regarding preoperative and postoperative course and early morbidity and mortality were recorded on a structured data collection form upon retrospective review of hospital charts. Post hospital discharge status was ascertained by mailed questionnaire or telephone interview. Relationships between categorical variables were examined with Fisher's exact test. Continuous variables are expressed as Median and lower and upper quartile values, or as mean±standard deviation. Differences between median values are evaluated with the Wilcoxon rank-sum test. Potential predicators of operative mortality were examined in univariate logistic regression models.

Given both the small number of operative deaths and the small number of patients in the study, we fit two-variable logistic regression models, each including presence of an abscess, to see if any second variable provided significant additional information. The same strategy was used to evaluate predictors of late mortality except Cox proportional hazards regression models were fit. Kaplan–Meier survival curves were compared with the logrank test.

Because of the small sample size and small number of deaths, failure to find significant predictors at the 0.05 level may result from lack of power, as well as from the lack or a relationship.

	3. Results

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 
The total follow-up was 584 patient-years with median follow up of 13 years. Native valve endocarditis was present in 40 patients (63%) and prosthetic valve endocarditis in 23 (37%). Active endocarditis (AE) was present in 38 patients (60%) and healed endocarditis (HE) in 25 patients (40%). The incidence of native valve endocarditis did not show any significant change over the 24-year period. In contrast, prosthetic valve endocarditis showed significant increase in relative incidence during the study period (from 4% during the period from 1977 to 1981 to 43% during the period from 1987 to 1991, P=0.03). Proportion of patients with active endocarditis increased significantly from 13% in the period from 1972 to 1976 to 29% during the period from 1992 to 1996 (P=0.04).

Streptococcus viridans was the most common causative organism for the entire population of patients with multivalve endocarditis (28%) with strong predilection for native valve endocarditis (42% for NVE versus 13% for PVE). Staphylococcus aureus and coagulase negative staphylococci were the most common causative microorganisms in the group of patients with PVE (31%). Culture-negative endocarditis occurred in 16 (25%) patients. Absence of positive blood cultures was more frequently observed in the group of patients with healed endocarditis than in patients with active endocarditis (48 vs. 18%, respectively, P=0.02). Incidence of vegetations was higher in NVE patients, with no difference in the incidence of abscess formation between the groups.

The preoperative echocardiographic evaluation was performed in 26 patients (42%). All patients underwent transthoracic echocardiography, with addition of transesophageal echocardiography in 14 patients (22%). The presence of vegetations was detected in 16 patients (61%).

Both duration of cardiopulmonary bypass time and aortic cross-clamp time were significantly longer in patients with prosthetic endocarditis (Table 1). Presence of active endocarditis was associated with significant increase in the duration of cardiopulmonary bypass and aortic cross-clamp time (Table 1). Patients who required aortic root reconstruction had significantly longer cardiopulmonary bypass times (201±93 vs. 137±85 min, P=0.02), and significantly longer aortic cross-clamp time (150±85 vs. 100±57 min, P=0.03), than patients who did not require root reconstruction.

A total of ten deaths resulted in an operative mortality of 16 percent. Of those, four deaths occurred in the native valve endocarditis group and six in prosthetic valve endocarditis group (P=0.15). The majority of deaths (seven) occurred in the patients with NYHA FC IV. The only statistically significant factor associated with operative death was the presence of abscess at the operation (P=0.03,) (Table 2). Postoperative stroke occurred in three patients (4%), in two patients strokes occurred prior to surgery.

View larger version (11K): [in this window] [in a new window]   Fig. 1. Kaplan–Meier survival curve for the multivalve surgery for endocarditis. Numbers above x-axis labels are the number of patients alive at the beginning of the time interval.

	4. Discussion

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 
This retrospective study presents the short and long term outcomes for the multivalve surgery for infectious multivalve endocarditis. The major predictor of early and late mortality is the presence of an abscess, which is the distinctive sign of the local advancement of disease.

Multivalve endocarditis is a separate clinical entity, which was recognized to be an independent risk factor affecting survival in patients with native or prosthetic endocarditis [7]. Our 24-year experience clearly shows an increase in the incidence of multivalve surgery for endocarditis in general population of patients undergoing valve surgery. Growing number of patients with prosthetic valves and relative resistance of prosthetic valve endocarditis to antimicrobial therapy resulted in a significant increase in the proportion of surgical patients with prosthetic valve endocarditis.

The optimal timing of surgery has been one of the main controversial issues in the management of infective endocarditis, especially in patients with involvement of more than one valve. Our results show the increasing number of patients in whom the surgery was performed in the active phase of endocarditis, reflecting the trend toward an early surgical intervention in the course of multivalve endocarditis, which corresponds with our previous findings from studies on single valve endocarditis [4,8].

The cure rates for both native and prosthetic valve endocarditis have been dependent on the type of the causative microorganism. Streptococcal endocarditis was the most frequent cause of native multivalve endocarditis in our patient population, which is consistent with the findings for single valve endocarditis. The most common cause of the prosthetic valve endocarditis in our study was Staphylococcus aureus and coagulase negative Staphylococcus. However, the presence of staphylococcal endocarditis was not identified to be a risk factor for early or late morbidity, which is in contrast to the prevailing findings from previous series [9,10]. Small number of patients with endocarditis associated with intravenous drug abuse, is reflective of overall small number of patients in the study and the fact that significant proportion of those patients present with the isolated tricuspid valve endocarditis. Relatively high proportion of culture negative patients in the healed endocarditis group is likely result of early initiation of antibiotic therapy in outside hospital prior to transfer to our institution, as well as unavailability of modern microbiology isolation techniques for our patients at the beginning of the study.

Although the estimated risk of systemic embolization rises with the number of affected valves, the overall incidence of perioperative strokes in our patients with multivalve endocarditis remained low. This may be attributed to the aggressive surgical approach with high proportion of patients operated in the early phase of endocarditis, thus preventing embolic complications.

Higher degree of valvular and annular destruction and preoperative hemodynamic instability can provide the explanation for the statistically significant prolongation of cardiopulmonary bypass time and aortic cross-clamp time in patients with prosthetic and active endocarditis. However, neither short- nor long-term outcomes of those patients were affected by the prolonged surgery. The explanation for those observations probably lies in the improvements in the cardiopulmonary bypass techniques and techniques of myocardial protection, which made even long and complex surgical procedures safe for the patients.

The prognosis in prosthetic valve endocarditis is clearly worse then in the native endocarditis [2,11,12]. The overall mortality for prosthetic valve endocarditis ranges form 30–80% for the early form and 20–40% for the healed form [12,13]. Annular abscess formation and extension of the infection into deep perivalvular tissue makes multivalve surgery technically more complex and strongly affects early mortality [13,14]. Every effort to prevent the local advancement of endocarditis and perform the multivalve replacement early in the course of the disease becomes a high priority worth exploring. The use of aortic and mitral homografts in the cases of annulus destruction may further improve the outcome in the complicated cases of multivalve endocarditis [15].

Application of mitral valve repair techniques in the treatment of native valve endocarditis offers significant advantage by allowing the restoration of valve competence without implantation of prosthetic material in contaminated valve tissue. One third of our patients underwent a valve repair as a part of their operation. Mid- and long-term results of mitral valve repairs in our patient population are encouraging. Aggressive forms of staphylococcal endocarditis in intravenous drug abusers, and high incidence of recurrent endocarditis are likely explanations for more frequent failure of tricuspid valve repairs. The extent of valvular and annular destruction are major limiting factors in attempts of valve reconstruction in patients with endocarditis.

The factors associated with late mortality following the surgery for endocarditis include the presence of prosthetic and early endocarditis, aortic valve involvement, emboli, congestive heart failure, and staphylococcal infection and renal failure [4,7,8,12]. In the setting of the multivalve endocarditis the presence o valvular abscess was the only identifiable risk factor for late mortality. Presence of vegetations, prosthetic valve endocarditis and staphylococcal as well as streptococcal infection showed association with late mortality but failed to reach statistical significance probably secondary to the small number of patients and relatively limited statistical power.

In conclusion, the early and late mortality of multivalve surgery for multivalve endocarditis is predominantly determined by the presence of deep tissue destruction by endocarditis. Therefore, aggressive preoperative diagnostic approach with routine use of transesophageal echocardiography should be applied in every patient with multivalve endocarditis. Early surgical intervention prior to the advancement of disease results in low early and late mortality and offers the best choice for cure of multivalve endocarditis.

	Footnotes

 
Presented at the 69th Scientific Sessions of the American Heart Association, New Orleans, Lousiana, USA, November 10–13, 1996.

	References

Top Abstract 1. Introduction 2. Material and methods 3. Results 4. Discussion References

 

1. Griffin M.R., Wilson W.R., Edwards W.D., O'Fallon W.M., Kurland L.T. Infective endocarditis: Olmsted county. Minnesota, 1950 through 1981. J Am Med Assoc 1985;254:1199-1202.[Abstract]
2. Bayer A.S. Infective endocarditis. Clin Infect Dis 1993;17:313-322.[Medline]
3. Vongpatanasin W., Hillis D.L., Lange R.A. Prosthetic heart valves. N Engl J Med 1996;335:407-416.[Free Full Text]
4. Aranki S.F., Santini F., Adams D.H., Rizzo R.J., Couper G.S., Kinchla N.M., Gildea J.S., Collins J.J., Cohn L.H. Aortic valve endocarditis: determinants of early survival and late morbidity. Circulation 1994;II:175-182.
5. Durack D.T., Lukes A.S., Bright D.K. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 1994;96:200-209.
6. Clark R.E., Edmunds L.H., Cohn L.H., Miller D.C., Weisel R.D. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Eur J Cardio-thorac Surg 1988;2:293-295.[Medline]
7. Grover F.L., Cohen D.J., Oprian C., Henderson W.G., Sethi G., Hammermeister K.E. Determinants of the occurrence of and survival from prosthetic valve endocarditis. J Thorac Cardiovasc Surg 1994;108:207-214.[Abstract/Free Full Text]
8. Aranki S.F., Adams D.H., Rizzo R.J., Couper G.S., Sullivan T.E., Collins J.J., Cohn L.H. Determinants of early mortality and late survival in mitral valve endocarditis. Circulation 1995;92(II):143-149.[Abstract/Free Full Text]
9. Agnihotri A.K., McGiffin D.C., Galbraith A.J., O'Brien M.F. The prevalence of infective endocarditis after aortic valve replacement. J Thorac Cardiovasc Surg 1995;110:1708-1720.[Abstract/Free Full Text]
10. Larbalestier R.I., Kinchla N.M., Aranki S.F., Couper G.S., Collins J.J.J., Cohn L.H. Acute bacterial endocarditis. Optimizing surgical results. Circulation 1992;8(5 Suppl):II68-II74.
11. Calderwood S.B., Swinski L.A., Karchmer A.W., Waternaux C.M., Buckley M. Prosthetic valve endocarditis: analysis of factors affecting outcome of therapy. J Thorac Cardiovasc Surg 1986;92:776-783.[Abstract]
12. Wolff M., Witchitz S., Chastang C., Régnier B., Vachon F. Prognostic factors of overall survival in a series of 122 cases and consequences for treatment decision. Chest 1995;108:688-694.[Abstract/Free Full Text]
13. Tornos P., Sanz E., Permanyer-Miralda G., Almirante B., Planes A.M., Soler-Soler J. Late prosthetic valve endocarditis: immediate and long term prognosis. Chest 1992;101:37-41.[Abstract/Free Full Text]
14. Rocchiccioli C., Chastre J., Lecompte Y., Gandjbakhch I., Gibert C. Prosthetic valve endocarditis: the case for prompt surgical management. J Thorac Cardiovasc Surg 1986;92:784-789.[Abstract]
15. Acar C., Tolan M., Berrebi A., Gaer J., Gouezo R., Marchix T., Gerota J., Chavaud S., Fabiani J.N., Deloche A., Carpentier A. Homograft replacement of the mitral valve. Graft selection, technique of implantation, and results in forty-three patients. J Thorac Cardiovasc Surg 1996;111:367-378.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. Lopez, A. Revilla, I. Vilacosta, E. Villacorta, C. Gonzalez-Juanatey, I. Gomez, M. J. Rollan, and J. A. San Roman Definition, clinical profile, microbiological spectrum, and prognostic factors of early-onset prosthetic valve endocarditis Eur. Heart J., March 2, 2007; 28(6): 760 - 765. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John G. Byrne Lawrence H. Cohn Sary F. Aranki Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mihaljevic, T. Articles by Aranki, S. F. Search for Related Content PubMed PubMed Citation Articles by Mihaljevic, T. Articles by Aranki, S. F. Related Collections Valve disease