(Circulation. 1995;92:143-149.)
© 1995 American Heart Association, Inc.
Articles |
From the Division of Cardiac Surgery, Brigham and Women's Hospital, and the Department of Surgery, Harvard Medical School, Boston, Mass.
Correspondence to Sary F. Aranki, MD, Division of Cardiac Surgery, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.
| Abstract |
|---|
|
|
|---|
Methods and Results Over a 24-year period, mitral valve surgery was performed in 96 patients for infective mitral valve endocarditis. Patient age ranged from 20 to 78 years (median age, 52 years). There were 44 women (46%), and 48 of the 96 patients (50%) were in New York Heart Association functional class IV before surgery. Native valve endocarditis (NVE) and prosthetic valve endocarditis (PVE) were present in 72 patients (75%) and 24 patients (25%), respectively. Surgery during the active phase of endocarditis (AE) was required in 60 patients (62%) and during the healed phase (HE) in 36 (38%). The main indications for surgery in the AE group were congestive heart failure (60%), active sepsis (67%), peripheral emboli (47%), and acute renal failure (20%), and for the HE group the main indication was progressive congestive heart failure (69%). The overall operative mortality was 5.2%. Multivariate logistic regression analysis identified PVE (odds ratio [OR] 22.5; ±95% confidence interval, CI, 1.9 to 268; P=.014) and an associated procedure (OR 13.3; ±95% CI, 1.5 to 120; P=.021) to be independent predictors for early mortality. Follow-up was 97% complete, with a median of 3.5 years. Overall 5- and 10-year survivals were 83±4% and 63±8%, respectively. Multivariate analysis for late mortality identified PVE to be a significant predictor of late mortality (hazards ratio=3.1, ±95% CI, 1.4 to 6.8, P=.006). There were no significant differences in long-term morbidity results among the various subsets of mitral valve endocarditis.
Conclusions Mitral valve surgery for infective endocarditis is a significant high-risk procedure for PVE and when combined with associated procedures. The activity of endocarditis does not appear to have any influence on early mortality or long-term survival.
Key Words: mortality mitral valve surgery prosthesis endocarditis
| Introduction |
|---|
|
|
|---|
The purpose of this retrospective study was to summarize and report our surgical experience with mitral valve endocarditis and its various subsets (NVE, PVE, AE, and HE) to determine their influence on early mortality and late survival.
| Methods |
|---|
|
|
|---|
Cardiopulmonary bypass, systemic hypothermia, and moderate hemodilution were used in all patients. Myocardial protection consisted of local and systemic hypothermias before 1976. Antegrade crystalloid cardioplegia has been used since that time with the addition of retrograde delivery of cardioplegia in 1990. Blood cardioplegia, both antegrade and retrograde, has been used increasingly over the last 5 years for reoperations. Mechanical and bioprosthetic valves were inserted with interrupted sutures. Mitral valve repair procedures involved a leaflet and an annular or a subannular technique with or without the use of an annuloplasty ring. Extensive débridement of vegetation and annular abscesses was performed. Abscess cavities were closed with a pericardial patch when it was necessary to restore disruption of atrioventricular continuity or when the closure was necessary to anchor the replacement device.
Early mortality and morbidity were defined as death or complications occurring within 30 days or during the same hospital stay after surgery. Endocarditis was labeled active if the patient required surgery before completion of a standard course of antibiotic treatment, the duration of which was variable and dependent on the severity of sepsis and the responsible microorganism, irrespective of whether there were ongoing signs of sepsis or whether the blood culture, surgical specimen, or both were positive for the infecting microorganism. Endocarditis was labeled healed if surgery was performed after completion of antibiotic treatment and met the aforementioned criteria.12 PVE was defined as infection occurring on any type of tissue or mechanical valve device including an annular ring. Early PVE was present if recurrent or residual endocarditis occurred within 60 days after surgery. Endocarditis occurring after 60 days was labeled late PVE. Culture negative endocarditis was present when no microorganisms could be identified on serial blood cultures in patients presenting with the clinical picture of endocarditis, particularly in the presence of a new regurgitant murmur, CHF, or the presence of vegetation on echocardiogram. These were confirmed by the presence of leaflet perforation, vegetation, and valvular or perivalvular tissue destruction. The presence of acute or chronic inflammatory changes confirmed the diagnosis of endocarditis microscopically.13 14
Data regarding hospitalization including perioperative morbidity and mortality were obtained from hospital records. Data on outcome after discharge and clinical status at follow-up were obtained by annual questionnaire and telephone interviews. Complications were confirmed by contacting the referring physician, by obtaining copies of the medical records, and by postmortem examination when available. Follow-up was 97% complete. The total follow-up time was 421 patient-years, median follow-up time was 42 months (range 1 to 257 months).
Statistical Analysis
Univariate analysis was performed with
the
2 test or Fisher's exact test.
Multivariate logistic regression analysis was
used to identify independent variables that predict early survival.
Late survival experiences of the different groups were compared by use
of the log rank test and by constructing Kaplan-Meier survival curves.
A combination of variables related to late survival, to freedom
from endocarditis, and to freedom from other events was examined using
a Cox proportional-hazards regression model.
| Results |
|---|
|
|
|---|
|
|
|
|
Microbiology
The distributions of the microorganisms for the
various
subsets of mitral valve endocarditis (AE, HE, NVE, or PVE) are shown in
Fig 3
. Streptococcus viridans (S. viridans),
culture negative, staphylococcus aureus (S. aureus), and
staphylococcus epidermidis (S. epidermidis) accounted for
the majority of cases with 24%, 23%, 17%, and 6%, respectively.
There was a trend for a marked decline in the incidence of
culture-negative endocarditis and a modest increase in the
incidence of S. aureus and S. viridans over the
years. S. viridans and S. aureus predominated
the AE group; culture-negative endocarditis predominated the NVE
and HE groups; and S. epidermidis and streptococcus group D
enterococcus predominated the AE and PVE groups. Early PVE occurred in
2 patients (8%) and late PVE in 22 patients (92%).
|
Operative Variables
The operative findings and variables are
shown in Table 3
for the AE and HE groups and in Table
4
for the NVE and PVE groups. The operative procedures in each subgroup
are also shown. Isolated mitral valve replacement was performed in 72
patients (80%). An associated procedure was performed in 19 patients
(20%) and included a CABG in 10, another valvular procedure in
3, and another procedure in 6.
|
|
Mortality and Morbidity
There were a total of 5 early deaths
for an overall operative
mortality (OM) of 5.2%. One of the deaths occurred in the NVE group,
for an OM of 1.4%; and 4 deaths occurred in the PVE group, for an OM
of 16.6% (P=.015 by Fisher's exact test). The OM for AE
was 2.8% compared with 6.7% for the HE group (P=.38 by
Fisher's exact test). Stepwise multivariate logistic
regression analysis identified PVE (OR 22.5; ±95 CI, 1.9 to
268, P=.014) and the presence of an associated procedure (OR
13.3; ±95% CI, 1.5 to 120; P=.021) to be an independent
predictor for early mortality. Variables examined for early
mortality with univariate analysis are listed in
the "Appendix."
Out of 91 survivors, there were 20 late
deaths, yielding an
overall late mortality of 22%, with overall 5- and 10-year survivals
of 83±4% and 63±8%, respectively (Fig 4
).
Survivals
at 5 and 10 years (Fig 5
) were 83±5% and 62±10%
for
AE and 85±7% and 71±11% for HE (P=.92),
respectively.
The 5- and 10-year survivals were 87±5% and 74±7% for NVE and
77±10% and 43±16% for PVE (P=.08), respectively.
Cox
proportional-hazards analysis identified PVE (hazards
ratio=3.1; ±95% CI, 1.4 to 6.8; P=.006) to be an
independent predictor for worse late survival.
|
|
Reoperation was required
in 14 patients (15%); of those 14, 9
surgeries (64%) were for structural valve degeneration, 2 (14%) for
infectious valve degeneration, and 3 (21%) for other causes. The log
rank test was used to calculate late survival experiences of the
different subgroups. Freedom from reoperation at 5 and 10 years was
92±4% and 62±13% for AE and 94±4% and 84±10 for HE
(P=.70), respectively. The 5- and 10-year freedom from
reoperation was 94±3% and 64±12% for NVE and 90±7% and
90±7%
for PVE (P=.33), respectively. The 5- and 10-year freedom
from infective valve degeneration (IVD) was 93±4% and 93±4% for
AE
and 100% and 89±10% for HE (P=.70), respectively,
96±3%
and 92±5% for NVE and 92±7 and 92±7 for PVE
(P=.8),
respectively. Kaplan-Meier curves for freedom from IVD are shown in Fig
6
.
|
| Discussion |
|---|
|
|
|---|
The activity of infective mitral valve endocarditis (active or healed) does not appear to have any influence on early mortality, late survival, or late morbidity. This contrasts with our previous findings11 where active endocarditis on the aortic valve whether native or prosthetic was a predictor for infectious valve degeneration or recurrent endocarditis whether early or late. A similar finding was reported by Grover et al10 who found a 0.8% per year incidence of PVE among 1032 patients randomized to receive a Bjork-Shiley spherical valve or a Hancock porcine heterograft. The most significant preoperative predictor of PVE was AE at the time of the initial operation (7.4% versus 0.9%; P=-.001). However, no separate analysis was performed to examine the mitral and aortic positions separately. They also found that the type of prosthesis (mechanical or bioprosthetic) had no influence on the risk of developing PVE. The reason for this disparity in the susceptibility to recurrent endocarditis between the aortic and mitral valves remains unclear. A possible explanation would be the significant proportion of the mitral valve patients who had a mitral valve repair. However, this was not an independent predictor in the multivariate analysis. Fuzellier et al16 recently reported on 35 patients operated on during the acute phase of mitral valve endocarditis. All patients underwent mitral valve repair using their techniques. The OM was 5.7%, and at a mean follow-up of 23 months no recurrence of endocarditis had been observed. One patient required reoperation and 3 died from noninfective related causes. Therefore, mitral valve repair in the setting of acute infective endocarditis is technically possible in some patients with good early and late results, minimizing morbidity rates associated with prosthetic valve devices.
Over the last 50 years, we have witnessed revolutionary and evolutionary changes in valvular endocarditis. The introduction of antibiotics resulted in a major modification of the clinical course and natural history of endocarditis.15 Patients no longer died of overwhelming sepsis but from CHF consequent to valvular tissue destruction.17 Another major development was the introduction of prosthetic valves in the early 1960s.18 19 A dramatic decline in the mortality associated with infective endocarditis occurred particularly in patients refractory to antibiotic therapy and in those with severe congestive heart failure.20 Accompanying these major developments that influenced the clinical course of endocarditis was the change in the population at risk.21 22 During this period there was a major decline in the incidence of rheumatic valvular heart disease.5 20 In addition, there was an increase in life expectancy23 accompanied by an increase in the prevalence of degenerative valvular heart disease.5 7 24 The increase in the number of drug addictionrelated valvular endocarditis cases9 21 25 and the emergence of more virulent and drug-resistant microorganisms9 26 were other factors that influenced the natural history of endocarditis. As a result of all these changes, more patients with acute or active endocarditis and more patients with PVE require surgical intervention.5 12
Mitral valve NVE is more common than aortic valve NVE and is more sensitive to antibiotic therapy, with a better chance for a medical cure.1 2 PVE on the other hand is more common in the aortic position than the mitral position.3 4 7 As a result, the overall incidence of surgical AVE exceeds that of the mitral valve. During the period of this study, only 4% of the total mitral valve procedures were required for mitral valve endocarditis. This compares with 6% total aortic valve procedures requiring surgery for aortic valve endocarditis during a 22-year period.11
The optimal timing of surgery has always been one of the main controversial issues in the overall management of infective valvular endocarditis. In the presence of any pressing need for acute surgical intervention such as repeated systemic emboli, large vegetation apparent on echocardiogram, and severe hemodynamic compromise with adverse multisystem effect, the decision should be obvious and immediate.27 28 29 However, the decision becomes somewhat more difficult in patients with active signs of sepsis and a somewhat delayed response to antibiotic therapy but with less severe symptoms. On the basis of the findings of this study, we conclude that the activity of endocarditis whether active or healed has no significant impact on short-term or long-term results. Therefore, it appears that the activity of endocarditis plays a minor role in the decision-making process regarding timing of surgery. Instead the severity of cardiac and extracardiac manifestation and the degree of hemodynamic stability along with the overall general condition of each individual patient becomes the major determinant of the optimal timing of surgery.
The surgical risks associated with NVE whether active or healed appear to have declined significantly over the years as shown by this and other studies.16 30 Nevertheless, the results with PVE continue to be dismal and associated with a significantly increased risk of early and late morbidity and mortality.15 29 31 32 33 34 35 36 37 38 Every effort to prevent its occurrence becomes a high priority worth exploring. As we have already mentioned, mitral valve repair even in the presence of active infection may reduce the incidence of recurrence on a prosthesis.16 39 Several recent reports have described the use of cryopreserved mitral homograft for the treatment of difficult and repeat endocarditis.40 41 However, these new techniques have had follow-up periods too short to allow one to draw any meaningful conclusions. Nevertheless, in the aortic position the use of homografts42 43 44 45 has reduced the incidence of recurrent endocarditis.
In conclusion, PVE on the mitral valve is an independent predictor of early mortality and late survival. The activity of the infection does not appear to influence the rate of recurrence of endocarditis. These results stress the need for continued, aggressive antibiotic prophylaxis in the presence of a mitral valve prosthesis.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Factors Examined for Early Mortality Using Univariate Analysis
Sex Age Year of surgery New York Heart Association
functional class Previous operation Active vs healed
endocarditis Native vs prosthetic endocarditis CHF Presence of
vegetation Preoperative embolism Acute renal failure Uncontrolled
sepsis Causative microorganism Perivalvular
leak Hemolysis Annular abscess Cusp perforation Associated
CABG Any associated procedure Low cardiac output
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. P. Remadi, G. Habib, G. Nadji, A. Brahim, F. Thuny, J. P. Casalta, M. Peltier, and C. Tribouilloy Predictors of Death and Impact of Surgery in Staphylococcus aureus Infective Endocarditis Ann. Thorac. Surg., April 1, 2007; 83(4): 1295 - 1302. [Abstract] [Full Text] [PDF] |
||||
![]() |
G Nadji, J P Remadi, F Coviaux, A Ali Mirode, A Brahim, M Enriquez-Sarano, and C Tribouilloy Comparison of clinical and morphological characteristics of Staphylococcus aureus endocarditis with endocarditis caused by other pathogens Heart, July 1, 2005; 91(7): 932 - 937. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Zegdi, M. Debieche, C. Latremouille, D. Lebied, C. Chardigny, J.-M. Grinda, S. Chauvaud, A. Deloche, A. Carpentier, and J.-N. Fabiani Long-Term Results of Mitral Valve Repair in Active Endocarditis Circulation, May 17, 2005; 111(19): 2532 - 2536. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Mihaljevic, J. G. Byrne, L. H. Cohn, and S. F. Aranki Long-term results of multivalve surgery for infective multivalve endocarditis Eur. J. Cardiothorac. Surg., October 1, 2001; 20(4): 842 - 846. [Abstract] [Full Text] [PDF] |
||||
![]() |
J M Guerra, M P Tornos, G Permanyer-Miralda, B Almirante, M Murtra, and J Soler-Soler Long term results of mechanical prostheses for treatment of active infective endocarditis Heart, July 1, 2001; 86(1): 63 - 68. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Jault, I. Gandjbakhch, A. Rama, M. Nectoux, V. Bors, E. Vaissier, P. Nataf, A. Pavie, and C. Cabrol Active Native Valve Endocarditis: Determinants of Operative Death and Late Mortality Ann. Thorac. Surg., June 1, 1997; 63(6): 1737 - 1741. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |