Abstract
Background:
Early surgery for infective endocarditis (IE) with acute heart failure (AHF) is recommended, despite clinical results being unclear. We investigated the effect of initial treatment in such patients.
Methods and Results:
Outcomes for 470 patients with active IE who underwent valvular surgery during 2009–2016 were reviewed. Of them, 177 had symptomatic AHF when diagnosed with IE (excluding those with cardiogenic shock or intubated for AHF). They were divided into 2 groups based on initial treatment: Group S (underwent valvular surgery immediately; n=74) and Group M (received initial medical treatment for infection and HF; n=103). The median (interquartile range) waiting period from diagnosis to surgery in Groups S and M was 1 (1–3) and 15 (8–33) days, respectively (P<0.001). The 5-year survival rate was higher in Group S than Group M (80% vs. 64%; P=0.108). Group M was divided into Group P (initial medical treatment was effective and elective surgery was performed; n=62) and Group E (emergency surgery was necessary during medical treatment; n=41); overall 5-year survival was significantly worse in Group E than Group P (42% vs. 79%; P<0.012). In Group M, multivariate analysis indicated that Staphylococcus aureus infection (odds ratio 3.82; 95% confidence interval 1.19–13.3; P=0.024) was a significant risk factor for conversion to emergency surgery.
Conclusions:
Considering poor outcomes of emergency surgery for medically refractory HF, early surgery may be a reasonable option for IE patients, especially those with S. aureus infection.
Despite recent advances in the diagnosis and treatment of infective endocarditis (IE), it remains a disease with a high mortality rate.1,2
Several studies have evaluated the risk factors for mortality in IE patients and found that acute heart failure (AHF) increases, whereas surgical intervention reduces, mortality.3,4
Although guidelines have generally recommended early surgery for IE patients with symptomatic AHF, the actual time for ‘early surgery’ remains unknown5–7
because clinical results in AHF patients belonging to a pure surgical cohort have never been evaluated. Although emergency surgery is required for patients with AHF from cardiogenic shock (CGS), whether non-cardiogenic AHF will be responsive to medical therapy cannot be judged at the time of IE diagnosis. Therefore, either surgical or medical treatment can be used for the initial management of IE patients with non-cardiogenic AHF, although the effect of the initial therapy on clinical results remains unknown. In the present study we analyzed surgical outcomes in IE patients complicated with AHF, and evaluated the effect of the initial therapy on prognosis.
Methods
Study Design and Patients
The present retrospective multicenter study was approved by the Institutional Review Board of Osaka University Hospital. The use of the Osaka Cardiovascular Research (OSCAR) Group database was approved by each affiliated hospital. Detailed records of patients who underwent valvular surgery for left-sided active IE at 12 affiliated hospitals since 2009 were obtained.8,9
The cohort in this study was equivalent to that in our previous study.8
From 2009 to 2016, 470 patients underwent valvular surgery for left-sided active IE at 12 hospitals. All patients satisfied Duke’s criteria.10
Of these patients, 57 (12%) were excluded from the study because they were in CGS or intubated for AHF, presenting definite indication for emergency surgery (Figure S1). Of the remaining 413 patients, we first evaluated the characteristics and clinical results of patients with and without symptomatic AHF. Second, we divided patients with AHF into 2 groups according to the initial intention-to-treat: Group M, who received initial medical treatment for AHF; and Group S, who underwent initial urgent surgery. Patient characteristics and clinical results were evaluated according to the initial intention-to-treat.
Definition of Active IE and AHF
Active IE in this study was defined as IE requiring intravenous administration of antibiotics until valvular surgery. AHF was defined as New York Heart Association Class IV, which is symptomatic at rest regardless of inotropic support at the time of IE diagnosis. Patients requiring only oral or intravenous diuretics without any symptoms of heart failure at rest were not regarded as patients with AHF.
Definition of Initial Intention-to-Treat
Initial intention-to-treat was defined as follows:
1. Initial urgent surgery (Group S) represented initial treatment immediately after consulting cardiac surgeons for urgent valvular surgery at the time of IE diagnosis, and the time of surgery fixed immediately after the consultation.
2. Initial medical treatment (Group M) represented initial treatment with antibiotics and medications for AHF, without any definite plan for valvular surgery at the time of IE diagnosis.
Patients in Group M were further subdivided in Group P (patients who proceeded to elective planned surgery after response to medical treatment) and Group E (patients who required conversion to emergency surgery because of worsening AHF).
Valve Surgery
The detail methods of valvular surgery for IE have been described previously.8
In the present study, a retrospective multicenter evaluation, the time of valvular surgery was decided by discussion between the cardiologist and surgeon at each hospital. Postoperative management was followed according to the guidelines of each hospital.6
Intravenous antibiotics were generally administered for 4 weeks postoperatively.
Data Collection, Echocardiographic Values, and Laboratory Data
Details of patient identification and data collection have been described previously.8
Echocardiographic parameters at the time of IE diagnosis were obtained. Follow-up investigations were performed using information obtained from medical records or by contacting the physician by telephone. Mid-term follow-up data were also collected in this study.
Definition of Complications During the Postoperative and Mid-Term Follow-up Periods
The definition of postoperative complications has been described previously.8
Mid-term mortality was defined as ‘death for any reason’. Cardiovascular-related death was defined as ‘death due to a cardiovascular disease’, including stroke, heart failure, sudden death, prosthetic valve endocarditis, and acute myocardial infarction. Cerebrovascular accident (CVA)-related death was defined as ‘death due to a major stroke’. Infection-related death was defined as ‘death due to any infection’ (e.g., recurrence of IE, pneumonia). Multiple organ failure (MOF)-related death was defined as ‘death due to MOF’. MOF-related deaths did not include deaths due to HF.
Statistical Analysis
Statistical analyses were performed using JMP version 13.0 (SAS Institute, Cary, NC, USA). Categorical variables are presented as percentages, whereas continuous data are presented as the mean±SD or as the median with interquartile range (IQR). For comparisons between 2 groups, the Chi-squared test was used for categorical variables; the significance of differences between continuous variables was evaluated using t-test or Wilcoxon test, as appropriate. All p-values for statistical analysis were 2-tailed and P<0.05 was considered significant.
Overall survival and freedom from recurrence of IE were estimated using Kaplan-Meier curves, and compared using a log-rank test (Figure 1). For comparison among more than 2 group characteristics, Bonferroni correction was performed for interpretation of statistical significance.
Cox regression hazard models were used to identify the risk factors for overall mortality. The multiple regression model was used to identify risk factors for the conversion to emergency surgery. Baseline characteristics, preoperative valuables, comorbidities, or complications were entered into the model for risk analysis of overall mortality. Univariate analysis was used initially, then factors with P<0.05 were included in the Cox hazard model to identify the risk factors for mortality. Parameters that showed significant differences on comparing Group P and Group E patients were used in the multiple regression model to identify risk factors for conversion to emergency surgery.
Results
Of the 413 patients, 236 (57%) had no symptomatic AHF at the time of IE diagnosis, whereas 177 (43%) had symptomatic AHF. Of the 177 patients with symptomatic AHF, 74 (42%) underwent urgent or emergency surgery as an initial intention-to-treat (Group S), whereas 103 (58%) underwent medical therapy (Group M; Figure S1). The preoperative characteristics of patients with and without AHF are compared in Table S1. Relative to patients without AHF, patients with AHF were significantly older and had more severe infection, impaired end-organ function, and worse echocardiographic parameters. Methicillin-resistant Staphylococcus aureus (MRSA) infection and aortic valve involvement were more prevalent in patients with AHF. Comparisons of patient characteristics and echocardiographic parameters at the time of IE diagnosis between Groups S and M are given in
Table 1. Importantly, although Group S had a higher white blood cell count than Group M, there were no significant differences in any other preoperative parameters, including comorbidities and complications, causative bacteria, valves affected, and echocardiographic parameters, between the 2 groups.
Table 1.
Patient Characteristics at the Time of Diagnosis of Endocarditis
| |
Group S
(n=74) |
Group M
(n=103) |
P value |
| Baseline characteristics |
| Age (years) |
60.7±16.7 |
64.2±14.3 |
0.135 |
| Male |
46 (48) |
65 (63) |
0.899 |
| Bacterial species |
| Positive blood culture |
55 (74) |
74 (72) |
0.716 |
| Staphylococcus species |
21 (28) |
38 (37) |
0.238 |
| Staphylococcus aureus |
12 (16) |
27 (26) |
0.115 |
| MRSA |
7 (9) |
11 (11) |
0.793 |
| Streptococcus species |
27 (36) |
30 (29) |
0.304 |
| Comorbidities |
| Atrial fibrillation |
7 (9) |
19 (18) |
0.097 |
| Hemodialysis |
6 (8) |
12 (12) |
0.445 |
| Diabetes mellitus |
12 (16) |
26 (25) |
0.151 |
| Valves affected |
| Aortic valve involvement |
40 (54) |
56 (54) |
0.967 |
| Mitral valve involvement |
44 (59) |
66 (64) |
0.535 |
| Aortic and mitral valve involvement |
10 (14) |
19 (18) |
0.385 |
| Prosthetic valve endocarditis |
10 (14) |
19 (18) |
0.385 |
| Echocardiographic parameters |
| LVDd (mm) |
55.0±7.8 |
55.5±8.5 |
0.717 |
| LVDs (mm) |
36.7±8.6 |
36.1±8.1 |
0.651 |
| Ejection fraction (%) |
61.0±12.0 |
63.3±10.3 |
0.205 |
| Systolic tricuspid pressure gradient (mmHg) |
39.0±14.6 |
40.4±16.5 |
0.665 |
| Maximum length of vegetation (mm) |
14.2±6.6 |
12.5±6.3 |
0.106 |
| Grade of aortic regurgitation |
1.7±1.6 |
1.9±1.7 |
0.468 |
| Grade of mitral regurgitation |
2.8±1.2 |
2.7±1.2 |
0.787 |
| Laboratory valuables at diagnosis |
| White blood cell count (×1,000/μL) |
12.5±5.7 |
10.9±4.9 |
0.076 |
| C-Reactive protein (mg/dL) |
7.3 [4.5–12.8] |
5.4 [3.2–10.5] |
0.344 |
| Hemoglobin (g/dL) |
10.1±2.0 |
10.2±1.8 |
0.766 |
| Platelet count (×1,000/μL) |
18.7±11.7 |
18.1±10.6 |
0.756 |
| Creatinine (mg/dL) |
0.94 [0.73–1.26] |
0.91 [0.71–1.87] |
0.121 |
| eGFR (mL/min/1.73 m2) |
84.3±43.8 |
75.8±44.0 |
0.227 |
| Total bilirubin (mg/dL) |
0.9 [0.6–1.2] |
0.7 [0.5–1.0] |
0.799 |
| Albumin (mg/dL) |
2.75±0.51 |
2.77±0.53 |
0.774 |
| BNP (pg/mL) |
696 [185–1,095] |
454 [215–638] |
0.780 |
| Other complications |
| Acute cerebral infarction |
25 (34) |
38 (37) |
0.672 |
| Acute intracranial hemorrhage |
5 (7) |
14 (14) |
0.149 |
| Systemic embolism |
30 (41) |
49 (48) |
0.356 |
Data are given as the mean±SD, n (%), or as the median [interquartile range]. BNP, B-type natriuretic peptide; eGFR, estimated glomerular filtration rate; Group M, patients who received initial medical treatment for infection and heart failure; Group S, patients who underwent valvular surgery immediately; LVDd, left ventricular end-diastolic dimension; LVDs, left ventricular end-systolic dimension; MRSA, methicillin-resistant Staphylococcus aureus.
Operative details, postoperative complications, and in-hospital mortality are given in
Table S2. AHF patients underwent valvular surgery at a median (IQR) of 6 (2–18) days after IE diagnosis. In AHF patients, the prevalence of mitral valve repair was lower, whereas the requirement for aortic valve annular reconstruction was higher. AHF patients required longer operative times, more blood products, and longer intubation periods and postoperative hospital stay. In-hospital mortality was significantly higher in patients with than without AHF (4% vs. 12%; P<0.001) because of a higher rate of MOF-related deaths in the former. Comparisons between Group S and Group M patients are given in
Table 2. Patients in Group S underwent valve surgery at a median (IQR) of 2 (1–3) days after IE diagnosis, compared with 15 (8–33) days in Group M. Reasons for patients receiving initial medical treatment were cerebral complications at the time of IE diagnosis (n=30; 29%), physicians believing that medical treatment would be effective (n=43; 42%), and patients being judged as being high-risk surgical patients (n=33; 32%). Importantly, 49 patients (48%) in Group M had initial medical treatment without consultation with cardiovascular surgeons because they were diagnosed in hospitals without a cardiovascular surgery department. Considering the details of valve surgery, there were no significant differences between Group S and Group M in terms of the prevalence of valve repair or replacement and the use of bioprosthetic or mechanical valves, but the requirement for reconstruction of the aortic annulus was more prevalent in Group M. There were no differences in postoperative complications and in-hospital mortality between the 2 groups.
Table 2.
Perioperative Characteristics
| |
Group S
(n=74) |
Group M
(n=103) |
P value |
| Days from diagnosis to surgery |
2 [1–3] |
15 [8–33] |
<0.001 |
| Valve surgery |
| Mitral valve |
n=47 |
n=69 |
0.633 |
| Mitral valve repair |
16 (34) |
27 (39) |
0.581 |
| Mitral valve replacement |
31 (66) |
42 (61) |
0.581 |
| Bioprosthetic valve |
23 (74) |
29 (69) |
0.637 |
| Mechanical valve |
8 (26) |
13 (31) |
0.637 |
| Aortic valve |
n=40 |
n=58 |
|
| Bioprosthetic valve |
32 (80) |
51 (88) |
0.289 |
| Mechanical valve |
8 (20) |
7 (12) |
0.289 |
| Aortic root surgery |
3 (8) |
6 (10) |
0.636 |
| Reconstruction of aortic annulus |
12 (16) |
29 (28) |
0.047 |
| Operative times and transfusion |
| Operative time (min) |
336±128 |
391±177 |
0.030 |
| CPB time (min) |
183±78 |
214±114 |
0.053 |
| Aortic clamp time (min) |
132±62 |
155±72 |
0.038 |
| Platelet transfusion (units) |
12.1±12.8 |
11.6±15.3 |
0.883 |
| Postoperative complications |
| Neurological complication |
17 (24) |
15 (15) |
0.130 |
| Acute intracranial hemorrhage |
5 (7) |
14 (14) |
0.147 |
| Mediastinitis |
2 (3) |
2 (2) |
0.739 |
| Requiring CVVHD |
8 (10) |
20 (19) |
0.123 |
| Complete AV block |
2 (3) |
2 (2) |
0.739 |
| Intubation duration (days) |
1 [1–4] |
1 [1–3] |
0.238 |
| Duration of postoperative intravenous ABx (days) |
29 [25–42] |
28 [18–40] |
0.211 |
| Hospital stay after surgery (days) |
41 [27–64] |
47 [31–59] |
0.590 |
| Mortality |
| 30-day mortality |
5 (7) |
9 (9) |
0.632 |
| Hospital mortality |
7 (9) |
15 (15) |
0.313 |
| Cause of deathA |
| Heart failure |
2 (3) |
3 (3) |
0.934 |
| Refractory infection |
2 (3) |
6 (6) |
0.327 |
| Cerebrovascular accident |
1 (1) |
4 (4) |
0.319 |
| Multiple organ failure |
5 (7) |
5 (5) |
0.591 |
Data are given as the mean±SD, n (%), or as the median [interquartile range]. AMore than 1 cause of death could be listed for each patient. ABx, antibiotics; AV, atrioventricular; CPB, cardiopulmonary bypass; CVVHD, continuous venovenous hemodialysis. Other abbreviations as in Table 1.
Overall survival rates at 1, 3, and 5 years were 79%, 70%, and 70%, respectively, in 177 AHF patients, compared with 91%, 85%, and 83%, respectively, in 236 patients without AHF (P=0.006;
Figure S2). Comparing the overall survival rates between Groups S and M revealed an overall survival rate at 1, 3, and 5 years of 86%, 80%, and 80%, respectively, in Group S, compared with 75%, 64%, and 64%, respectively, in Group M (P=0.108). Overall survival was significantly better in Group S than in Group M for patients with aortic valve involvement, whereas there were no differences in survival for patients with mitral valve involvement (Figure 2B,C).
Postoperative neurological deterioration in Group S included cerebral deterioration in 2 (8%) of 25 infarction patients and in 0 (0%) of 5 hemorrhage patients, compared with 1 (2%) of 49 non-infarction patients and 3 (4%) of 69 non-hemorrhage patients. There were no significant differences in any cerebral events after surgery in Group S regardless of the presence of cerebral complications before surgery.
Of 103 Group M patients, 62 (60%) underwent elective valve surgery after effective medical treatment a median of 22 days after IE diagnosis (Group P), whereas 41 (40%) required conversion to emergency surgery because of deteriorating AHF a median of 9 days after diagnosis (Group E). Patient characteristics, operative details, and operative results were compared between Groups P and E (Table 3). Although patients in Group E had a higher prevalence of S. aureus infection and a lower prevalence of mitral valve involvement and larger vegetation size, other parameters at the time of IE diagnosis were similar between the 2 groups. Considering preoperative laboratory parameters, higher serum C-reactive protein and bilirubin concentrations were seen in Group E. Nineteen patients (46%) in Group E required inotropic support before surgery. Although there were no differences in in-hospital mortality between Groups P and E (13% vs. 20%, respectively; P=0.369), patients in Group E had a longer hospital stay after surgery. Moreover, the prevalence of patients who could be discharged home was significantly lower in Group E than Group P (48% vs. 81%, respectively; P=0.001) and more than half the patients in Group E were transferred to long-term rehabilitation (19% vs. 52% for Groups P and E, respectively; P=0.001). Overall survival was significantly worse in Group E than in Groups P and S (Figure 3A). This worse survival rate in Group E was seen in patients who were discharged or transferred after surgery (Figure 3B). Overall survival was also compared among Groups S, P, and E for those with aortic or mitral valve IE patients (Figure S2). We further analyzed freedom from IE recurrence among Groups S, P, and E and found it to be almost similar among the 3 groups (Figure S3).
Table 3.
Comparison Between Patients With Planned Surgery (Group P) and Patients With Conversion to Emergency Surgery (Group E)
| |
Group P
(n=62) |
Group E
(n=41) |
P values |
| Baseline characteristics |
| Age (years) |
63.0±14.8 |
66.0±13.8 |
0.301 |
| Male |
37 (60) |
28 (68) |
0.198 |
| Bacterial species |
| Positive blood culture |
41 (66) |
33 (80) |
0.124 |
| Staphylococcus species |
19 (31) |
19 (46) |
0.144 |
| Staphylococcus aureus |
10 (16) |
17 (41) |
0.006 |
| MRSA |
6 (10) |
5 (12) |
0.750 |
| Streptococcus species |
21 (34) |
9 (22) |
0.268 |
| Comorbidities |
| Hemodialysis |
4 (6) |
8 (20) |
0.060 |
| Diabetes mellitus |
13 (21) |
13 (32) |
0.252 |
| Antiplatelet therapy |
12 (19) |
5 (12) |
0.298 |
| Cerebral complication |
14 (23) |
16 (37) |
0.128 |
| Valves affected |
| Aortic valve involvement |
33 (57) |
25 (61) |
0.437 |
| Mitral valve involvement |
44 (71) |
25 (61) |
0.094 |
| Aortic and mitral valve involvement |
16 (26) |
9 (22) |
0.621 |
| Prosthetic aortic valve endocarditis |
13 (21) |
6 (15) |
0.412 |
| Echocardiographic parameters |
| Ejection fraction (%) |
63.4±10.6 |
63.1±10.1 |
0.898 |
| Systolic tricuspid pressure gradient (mmHg) |
42.2±18.1 |
37.1±13.1 |
0.252 |
| Maximum length of vegetation (mm) |
11.4±6.7 |
13.9±5.7 |
0.076 |
| Laboratory valuables at diagnosis |
| White blood cell count (×1,000/μL) |
10.9±5.1 |
11.1±4.7 |
0.772 |
| C-Reactive protein (mg/dL) |
5.9 [3.9–10.3] |
5.3 [2.9–9.3] |
0.335 |
| Platelet count (×1,000/μL) |
17.6±9.8 |
18.8±11.9 |
0.619 |
| eGFR (mL/min/1.73 m2) |
74.4±40.7 |
77.7±48.3 |
0.729 |
| Total bilirubin (mg/dL) |
0.70 [0.5–1.0] |
0.75 [0.5–1.0] |
0.432 |
| Laboratory valuables just before surgery |
| White blood cell count (×1,000/μL) |
9.7±5.6 |
6.7±3.1 |
0.922 |
| C-Reactive protein (mg/dL) |
1.5 [0.7–4.4] |
4.0 [1.6–6.8] |
0.003 |
| Platelet count (×1,000/μL) |
18.8±9.6 |
18.2±9.4 |
0.749 |
| eGFR (mL/min/1.73 m2) |
80.0±47.4 |
80.5±51.3 |
0.957 |
| Total bilirubin (mg/dL) |
0.6 [0.4–0.8] |
0.7 [0.5–1.1] |
0.045 |
| Operative times and transfusion |
| Operative time (min) |
389±172 |
394±189 |
0.905 |
| CPB time (min) |
213±117 |
216±113 |
0.893 |
| Aortic clamp time (min) |
152±68 |
159±78 |
0.671 |
| Platelet transfusion (units) |
0 [0–20] |
10 [0–30] |
0.060 |
| Postoperative complications |
| Neurological deterioration |
3 (5) |
2 (5) |
1.000 |
| Intracranial hemorrhage |
2 (3) |
4 (9) |
0.205 |
| Mediastinitis |
0 (0) |
2 (5) |
0.156 |
| Requirement of CVVHD |
10 (16) |
10 (24) |
0.320 |
| Complete AV block |
0 (0) |
2 (5) |
0.156 |
| Intubation duration (days) |
1 [1–2] |
2 [1–6] |
0.155 |
| Post-operative intravenous ABx duration (days) |
28 [17–40] |
28 [21–40] |
0.457 |
| Hospital stay after valve surgery (days) |
36 [25–61] |
52 [34–70] |
0.089 |
| In-hospital mortality |
8 (13) |
8 (20) |
0.369 |
| Discharge home after treatment |
44 (81) |
16 (48) |
0.001 |
| Transfer to another hospital after treatment |
10 (19) |
17 (52) |
0.001 |
Data are given as the mean±SD, n (%), or as the median [interquartile range]. Abbreviations as in Tables 1,2.
The risk factors for overall mortality in 177 AHF patients were analyzed using a Cox hazard regression model (Table 4), which identified age, MRSA infection, preoperative hemodialysis, diabetes, prosthetic valve endocarditis, larger left ventricular end-diastolic dimension, lower platelet counts, and conversion to emergency surgery as risk factors. The choice of the initial intention-to-treat strategy was not a risk factor for overall mortality. Multivariate analysis revealed prosthetic valve endocarditis (hazard ratio [HR] 2.83; 95% confidence interval [CI] 1.29–5.88; P=0.011) and conversion to emergency surgery (HR 2.62; 95% CI 1.34–5.12; P=0.005) as independent risk factors for mortality. To determine the reasons for worse outcomes in Group E patients, we further divided the causes of death and compared them between Groups P and E. The rates for freedom from cardiovascular-, CVA-, infection-, and MOF-related deaths are shown
Figure S4. In this subanalysis, Group E patients had worse results for cardiovascular-, CVA-, and organ failure-related deaths.
Table 4.
Cox Hazard Analysis of Overall Mortality in 177 Patients With Acute Heart Failure
| |
Univariate analysis |
Multivariate analysis |
| HR (95% CI) |
P value |
HR (95% CI) |
P value |
| Age (years) |
1.03 (1.01–1.06) |
0.006 |
1.02 (1.00–1.05) |
0.090 |
| MRSA |
3.48 (1.57–6.91) |
0.003 |
1.93 (0.72–4.68) |
0.165 |
| Hemodialysis |
2.26 (1.02–4.47) |
0.045 |
2.14 (0.83–5.04) |
0.096 |
| Diabetes mellitus |
2.12 (1.11–3.87) |
0.023 |
2.02 (1.00–4.00) |
0.051 |
| Prosthetic valve endocarditis |
2.15 (1.07–4.03) |
0.034 |
2.83 (1.29–5.88) |
0.011 |
| LVDd (mm) |
0.95 (0.91–0.99) |
0.013 |
0.98 (093–1.02) |
0.313 |
| Platelet count (×1,000/μL) |
0.96 (0.93–0.99) |
0.024 |
0.98 (0.93–1.02) |
0.207 |
| Initial medical treatment |
1.55 (0.85–2.94) |
0.156 |
|
|
| Conversion to emergency surgery |
2.84 (1.57–5.06) |
0.001 |
2.62 (1.34–5.12) |
0.005 |
CI, confidence interval; HR, hazard ratio. Other abbreviations as in Tables 1.
Because conversion to emergency surgery was revealed as 1 of the independent risk factors in heart failure patients, we further evaluated this in 103 Group M patients (Table 5). Univariate analysis revealed that S. aureus infection, preoperative hemodialysis, and higher C-reactive protein concentrations at diagnosis were predictors for conversion to emergency surgery. Subsequent multivariate analysis revealed S. aureus infection as the only independent predictor for conversion to emergency surgery.
Table 5.
Risk Analysis for Conversion to Emergency Surgery in 103 Patients With Intention-to-Treat Initial Medical Therapy
| |
Univariate analysis |
Multivariate analysis |
| OR (95% CI) |
P value |
OR (95% CI) |
P value |
| Staphylococcus aureus |
3.68 (1.49–9.51) |
0.005 |
3.82 (1.19–13.3) |
0.024 |
| Length of vegetation (mm) |
1.07 (0.99–1.15) |
0.072 |
1.06 (0.98–1.16) |
0.131 |
| Hemodialysis |
3.52 (1.03–14.0) |
0.045 |
1.21 (0.20–6.52) |
0.826 |
| C-reactive protein (mg/dL) |
1.09 (1.03–1.17) |
0.002 |
1.04 (0.96–1.13) |
0.370 |
| White blood cell count (×103/mm3) |
1.17 (1.06–1.30) |
0.001 |
1.06 (0.93–1.21) |
0.402 |
CI, confidence interval; OR, odds ratio.
Discussion
Early surgery has been recommended for patients with symptomatic AHF (Class I, Level B).5,6
In the European Society of Cardiology guidelines, urgent surgery is recommended for patients with medically refractory AHF or echocardiographic signs of poor hemodynamic tolerance, although it is sometimes impossible to evaluate this at the time of IE diagnosis. In fact, in the present multicenter study, patient characteristics and parameters during IE diagnosis were almost similar between Groups S and M. This similarity could imply that the initial treatment was decided by each physician, and there is no definite threshold or consensus about initial treatment in ‘non-cardiogenic’ AHF patients. The American Association of Thoracic Surgery Guidelines also mention that deciding the real time of surgery for IE needs further evaluation.7
In the present study, surgery appears to be superior to medical therapy as an initial therapy, even in IE patients with ‘non-cardiogenic’ AHF. This superiority was particularly seen in patients with aortic valve involvement. Regarding aortic valve IE, vegetation easily destroys the valve leaflets and annulus compared with mitral valve IE and can cause hemodynamic collapse. Destruction of the aortic annulus eventually requires more complicated surgery.8,11
Regarding mitral valve IE, there was no superior possibility of mitral valve repair even if patients could proceed to elective surgery after effective medical therapy (Group P) compared with patients who had initial valvular surgery (Group S). The clinical results of patients who required conversion to emergency surgery for worsening AHF have never been reported. In this study, we demonstrate that conversion to emergency surgery impaired overall survival, especially survival after discharge or transfer to other hospitals. Moreover, Group E patients had longer hospital stays after valvular surgery, and the prevalence of patients who could be discharged was significantly lower (more than half the Group E patients were transferred to other hospitals for further treatment or rehabilitation). These results imply that the condition of Group E patients was worse at discharge because of longer waiting periods with refractory AHF or worse preoperative condition. Therefore, early identification of patients unresponsive to initial medical treatment is indispensable, along with the identification of patients who cannot proceed to initial surgical treatment (e.g., patients with hemorrhagic stroke).
To avoid conversion to emergency surgery, it is also important to know the risk factors for conversion. In this study, the presence of S. aureus was the exclusive risk factor, which is also a well-known risk factor for mortality.1,12,13
S. aureus
infection is more aggressive in destroying valvular structures, even during appropriate antibiotic therapy. In the present study, 17 of 27 (63%) Group M patients with S. aureus infection required conversion to emergency surgery. In addition to the preoperative severity of illness due to S. aureus- associated IE,14
high mortality in patients with Staphylococcus infection may be explained by this high rate of conversion to emergency surgery.
The present study has several limitations. First, the study was a retrospective multicenter study with limited patients in each hospital. Although the basic strategy was managed according to the guidelines, there was a possibility of differences in treatment strategy (e.g. valve repair or replacement, antibiotic duration). Second, although IE with AHF is an indication for surgery, it was possible that only medical therapy was used in some patients because the patients’ symptoms improved with medication or patients were too sick and estimated as being of high-surgical risk due to several complications, such as cerebral complications. Those patients were excluded from the analysis.
Conclusions
In conclusion, initial surgical treatment is superior to medical treatment in IE patients with non-cardiogenic AHF. Considering poor outcomes in patients who required conversion to emergency surgery for medically refractory AHF, early surgery should be considered for every IE patient with AHF, especially those who present with S. aureus infection.
Conflict of Interest
None declared.
Supplementary Files
Supplementary File 1
Figure S1.
Disposition of patients in the present study.
Figure S2.
Overall survival after valve surgery (A: patients with aortic valve involvement,
B: patients with mitral valve involvement) for patients who underwent initial surgery (Group-S, blue), and those who underwent planned elective surgery after medical treatment (Group-P, green), and those who required conversion to emergent surgery (Group-E, red).
Figure S3.
Freedom of recurrence after valve surgery in patients who underwent initial surgery (Group-S, blue), and those who underwent planned elective surgery after medical treatment (Group-P, green), and those who required conversion to emergent surgery (Group-E, red).
Figure S4.
Event-free survival for (A) cardiovascular-, (B) cerebrovascular accident-, (C) infection-, and (D) organ failure-related deaths in patients who underwent planned elective surgery after medical treatment (Group P) and those who required conversion to emergency surgery (Group E).
Table S1.
Characteristics at the time of diagnosis of endocarditis of patients with and without acute heart failure (AHF+ and AHF−, respectively)
Table S2.
Operative results in Patients with and without acute heart failure (AHF+ and AHF−, respectively)
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-18-0510
References
- 1.
Park LP, Chu VH, Peterson G, Skoutelis A, Lejko-Zupa T, Bouza E, et al. Validated risk score for predicting 6-month mortality in infective endocarditis. J Am Heart Assoc 2016; 5: e003016.
- 2.
Bae IG, Federspiel JJ, Miro JM, Woods CW, Park L, Rybak MJ, et al. Heterogeneous vancomycin-intermediate susceptibility phenotype in bloodstream methicillin-resistant Staphylococcus aureus isolates from an international cohort of patients with infective endocarditis: Prevalence, genotype, and clinical significance. J Infect Dis 2009; 200: 1355–1366.
- 3.
Chu VH, Park LP, Athan E, Delahaye F, Freiberger T, Lamas C, 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.
- 4.
Kiefer T, Park L, Tribouilloy C, Cortes C, Casillo R, Chu V, et al. Association between valvular surgery and mortality among patients with infective endocarditis complicated by heart failure. JAMA 2011; 306: 2239–2247.
- 5.
Habib G, Lancellotti P, Antunes MJ, Bongiorni MG, Casalta JP, Del Zotti F, 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.
- 6.
Baddour LM, Wilson WR, Bayer AS, Fowler VG Jr, Tleyjeh IM, Rybak MJ, 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.
- 7.
Pettersson GB, Coselli JS, Writing C, Pettersson GB, Coselli JS, Hussain ST, 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.e29.
- 8.
Yoshioka D, Toda K, Yokoyama JY, Matsuura R, Miyagawa S, Shirakawa Y, et al. Recent surgical results for active endocarditis complicated with perivalvular abscess. Circ J 2017; 81: 1721–1729.
- 9.
Yoshioka D, Toda K, Okazaki S, Sakaguchi T, Miyagawa S, Yoshikawa Y, et al. Anemia is a risk factor of new intraoperative hemorrhagic stroke during valve surgery for endocarditis. Ann Thorac Surg 2015; 100: 16–23.
- 10.
Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: Utilization of specific echocardiographic findings. Am J Med 1994; 96: 200–209.
- 11.
Graupner C, Vilacosta I, SanRomán J, Ronderos R, Sarriá C, Fernández C, et al. Peri-annular extension of infective endocarditis. J Am Coll Cardiol 2002; 39: 1204–1211.
- 12.
Pericart L, Fauchier L, Bourguignon T, Bernard L, Angoulvant D, Delahaye F, et al. Long-term outcome and valve surgery for infective endocarditis in the systematic analysis of a community study. Ann Thorac Surg 2016; 102: 496–504.
- 13.
Pang PY, Sin YK, Lim CH, Tan TE, Lim SL, Chao VT, et al. Surgical management of infective endocarditis: An analysis of early and late outcomes. Eur J Cardiothorac Surg 2015; 47: 826–832.
- 14.
Han SM, Sorabella RA, Vasan S, Grbic M, Lambert D, Prasad R, et al. Influence of Staphylococcus aureus on outcomes after valvular surgery for infective endocarditis. J Cardiothorac Surg 2017; 12: 57.
