Presentaci

1
Carlos-A. Mestres, MD, PhD, FETCS Consultant Card
iovascular Surgery Hospital Clínico. University
of Barcelona Barcelona. Spain
2
Infective endocarditis is an uncommon
disease associated to significant morbidity and
mortality. As in any infection within the
cardiovascular surgery, early diagnosis and
aggresive management are indicated
Infective endocarditis is a medical surgical
disease which must be managed by a
multidisciplinary team with shared interests
3
The Team
The Hospital Clinico of Barcelona Endocarditis
Study Group is a multidisciplinary group
specifically dedicated to the study and
treatment of infective endocarditis and
cardiovascular infections operational for 25
years
Infectious Diseases (6), Cardiovascular Surgery
(3), Microbiology (3), Surgical Pathology (1),
Echocardiography (2)
4
The Team
Infectious Diseases J.M.Miró, A.Moreno, A.
Del Río, N. De Benito, X.Claramonte,
J.P.Horcajada Cardiovascular Surgery
C.A.Mestres, R.Cartañá, S.Ninot, J.L.Pomar
Microbiology M.Almela, F.Marco, C.García
Surgical Pathology J.Ramírez, N.Pérez
Echocardiography J.C.Paré, M.Azqueta, M.Sitges
5
Infective Endocarditis
What have we learned? What have we changed? What
are we doing? Where are we going?
An overview
6
A - Short Courses of Therapy for Infective
Endocarditis
B - Infective Endocarditis in Drug Abusers
(IVDAs)
C Surgical experience
7
Potential number of candidates for short-courses
of therapy for right-sided MSSA endocarditis in
IVDAs at the Hospital Clínic of Barcelona, Spain
(1979-98)
Types of endocarditis in IVDAs
MSSA N ()
N
- Right-sided IE - Left-sided IE - Mixed IE
Total
142 46 16 204
104 (73) 16 (35) 10 (64) 130 (64)
2 wk Tx 40
According to methicillin-susceptibility, HIV
status and CD4 cell counts (gt200/µL)
8
Short Courses of Therapy for Infective
Endocarditis CONCLUSIONS
5. Patients allergic to penicillin who must
receive vancomycin with or without an
aminoglycoside must be treated during 4 wks
6. In our 25-year experience, one of every five
episodes of native valve IE (general population
IVDAs) and almost one of every two episodes of IE
in IVDAs were considered potential candidates for
these short courses (2 wks) of therapy
9
Infective Endocarditis in IVDAs HIV
infection SUMMARY
1. The incidence of IE in IVDA in the AIDS era is
decreasing probably due to the change of the drug
administration habits in order to avoid
HIV-infection
2. HIV-infected IVDA have a higher ratio of
right-sided IE and S. aureus endocarditis than
HIV-negative IVDA with IE
3. Mortality between HIV-infected or
non-HIV-infected IVDA with IE is similar.
However, mortality among HIV-infected IVDA is
higher in IVDA with less than 200 CD4 cells/µL
or with AIDS criteria
10
Infective Endocarditis in IVDAs HIV
Infection SUMMARY
4. IVDA with non-complicated MSSA right-sided IE
can be succesfully treated with an IV
short-course regimen of nafcillin or cloxacillin
plus an aminoglycoside during 2 weeks, although
the addition of an aminoglycoside may be avoided
or reduced to the first 3-7 days
5. Tricuspid valve replacement using mitral
homografts can be a safely alternative to
tricuspid valvulectomy for those IVDA with
endocarditis who need right heart surgery
Long-term results after cardiac surgery in
patients infected with the human immunodeficiency
virus type-1 (HIV-1) Mestres CA et al. Eur J
Cardio-thorac Surg 2003 231007-1016
11
Epidemiology
1990 - 2000
Diagnosis of IE 421 IV (IVDA) drug
abuse 104 General population 317 Native
IE 213 PVE 75 Pacemaker/AICD
29 Admissions/yr gt50
12
Infective endocarditis in intravenous drug
abusers and HIV-1 infected patients J.M.Miró, A.
del Río, C.A.Mestres Infect Dis Clin North Am
2002 16273-295
Infective endocarditis not related to
intravenous drug abuse in HIV-1-infected
patients report of eight cases and review of the
literature J.E.Losa, J.M.Miró, A. Del Río,
A.Moreno-Camacho, F.Gracia, X.Claramonte,
F.Marco, C.A.Mestres, M.Azqueta, J.M.Gatell and
the Hospital Clinic Endocarditis Study Group Clin
Microbiol Infect 2003 945-54
Surgical treatment of pacemaker and
defibrillator lead endocarditis. The impact of
electrode lead extraction on outcome A.del Río,
I.Anguera, J.M.Miró, L.Mont, Fowler VG Jr,
M.Azqueta, C.A.Mestres and the Hospital Clinic
Endocarditis Study Group Chest 2003
1241451-1459
13
NVE 387 - ADVP 237 - PVE 130 - PM 49 - All 803
14
PVE 132
15
S.aureus 274
16
ICE
Presumed intravascular catheter source by region
International Collaboration on Endocarditis
17
Specific indications
Mechanical valve Young, good ring, cured
IE Bioprosthesis Elderly (?), good ring,
cured IE Homograft Complicated IE, abscess,
annular destruction
18
The complicated root

1. Root abscess
2. Aorto-cavitary fistula

19
Aorto-cavitary fistulae
20
L770 - AORTO-CAVITARY FISTULIZATION IN
COMPLICATED ENDOCARDITIS. CLINICAL AND
ECHOCARDIOGRAPHIC FEATURES OF 76 CASES
(1992-2001) AND PROGNOSTIC FACTORS OF MORTALITY
42nd ICAAC. San Diego, CA. September 27-30, 2002

• The Spanish Aorto-cavitary Fistula Endocarditis
  Working Group

21
No clinical infective endocarditis (IE) series
have been performed studying the development of
aorto-cavitary fistulas (ACF) as a result of
spread of infection from valvular tissue towards
perivalvular structures. Our aims were to
investigate the clinical, echocardiographic and
microbiologic features and prognostic factors of
in-hospital mortality in patients with IE and
ACF. Retrospective and multicentre study at 11
Spanish and 1 North-american Hospitals in
patients with IE and ACF.
22
Basic considerations

• Spread of infection in infective endocarditis
  (IE) from valvular structures to the surrounding
  perivalvular tissue results in periannular
  complications.
• Rupture of abscesses and pseudoaneurysms in the
  sinuses of Valsalva result in the development of
  aorto-cavitary fistulas and intracardiac shunts.
• Aorto-cavitary fistula formation is an unusual
  complication of IE. An incidence of 1 of all
  cases of IE has been estimated. Fistulization of
  perivalvular abscesses occurs in 6-9 of cases.

23
Multicenter, international, retrospective,
descriptive study performed between 1992 and
2001 Infective endocarditis diagnosed according
to Duke criteria Aorto-cavitary fistulization
documented by TTE/TEE Univariate analysis of
prognostic factors of mortality
24
ACF n Cases IE n Incidence
General population Native valve Aortic Mitral
Other PVE Aortic Mitral Other Pacemaker
IV Drug abusers OVERALL
69 38 38 -- -- 31 31 -- -- -- 7 76
3147 2105 1056 930 119 872 536 326
10 170 1534 4681
2.2 1.8 3.6 --- --- 3.5 5.8 --- --- --- 0.4 1.6
25
Clinical characteristics
NVE45 PVE31 All76
Mean age (y) Male gender Previous valve
disease Comorbidity Mechanical ventilation IV
drug abuse Duration of symptoms (d) Duration to
Dx of ACF (d) CHF Neuro events Renal
failure Peripheral emboli Complete AV block
50.918.7 36 (80) 13 (28) 18 (40) 6 (13)
7 (16) 24.518.7 36.231.6 31 (69) 8 (18) 20
(44) 8 (18) 5 (11)
60.213.4 20 (65) 31 (100) 9 (29) 1 (3)
0 29.837.7 44.155.5 16 (52) 4 (13) 8
(26) 7 (23) 6 (19)
54.717.2 56 (74) 44 (59) 27 (36) 7 (9) 7
(9) 26.727.9 39.442.8 47 (62) 12 (16) 28
(37) 15 (20) 11 (14)
26
Pathogens
NVE45 PVE31 All76
Staphylococcus spp S.aureus CNS Streptococcus
spp VGS S.bovis Other streptococci Enterococcus
spp Culture negative Other (HACEK)
17 (38) 13 (29) 4 (9) 16 (35) 10
(22) 2 (4) 4 (9) 2 (4) 5 (11)
7 (15)
18 (58) 3 (10) 15 (48) 9 (29) 5
(16) -- 4 (13) 2 (6) -- 2 (6)
35 (46) 16 (21) 19 (25) 25 (33) 15 (20) 2
(3) 8 (10) 4 (5) 5 (6) 9 (12)
NVE vs PVE groups (plt0.05)
27
Echocardiography
Diagnostic yield of TTE and TEE
TTE n () TEE n ()
Native valve 26/44 (59) 31/33 (94)
PVE 15/31 (48) 28/28 (100)
Overall 40/75 (53) 59/61 (97)
28
Native N45
Prosthetic N31
Total N76
Echo findings
Patients with vegetations Mean maximal veg. size
(mm) Vegetations gt 10 mm Patients with abscess
Mean maximal abscess diameter Abscess gt 10
mm Ventricular septal defect Mean EF
() Mean LVEDD (mm) Multivalvular infection
83 11.7 56 78 12 mm 54 20 61.7 54.9 30

96 11.5 49 71 10 mm 44 21
62.5 55.2 33
65 12.1 70 87 15 mm 67 19
60.5 54.4 26
Native vs prosthetic, p lt 0.05
29
Total N76
Native N45
Prosthetic N31
Echo findings

• Fistulized sinus of Valsalva (SV)
• Right SV
• Left SV
• Non coronary SV
• Fistulized cardiac chamber ()
• Right atrium
• Right ventricle
• Left atrium
• Left ventricle
• Multiple
• Moderate/severe regurgitation

37 38 25 17 25 26 16 12 49
44 35 20 18 31 22 13 11 64
26 42 32 16 16 32 19 13 26
Native vs prosthetic, p lt 0.05
30
Total N76
Native N45
Prosthetic N31

• Surgical treatment
• Time to surgery
• lt 24 hours
• 2 - 7 days
• gt 7 days
• Closure of fistula ()
• Simple
• Pericardial patch
• Gore-tex patch
• Valve replacement
• Bioprosthesis
• Mechanical
• Homograft

87 24 42 34 41 48 11 92 24 50 18
87 33 36 31 41 46 13 95 28 49 18
87 11 52 37 41 52 7 89 19 52 19
31
Total N76
Native N45
Prosthetic N31

• In-hospital mortality
• - Surgical group (N66)
• - Medical group (N10)

31 (41) 28 (42) 3 (30)
16 (36) 13/39 (33) 3/6 (50)
15 (48) 15/27 (55) 0/4 (-)
Medical N3
Surgical N28
Cause of death - Multiorgan failure - Sudden
death - Septic shock - Cardiogenic shock -
Hemorrhage
23 10 26 19 23
33 33 - 33 -
32
Medical N7
Surgical N38

• Lost for follow-up
• Follow-up (mo., mean, range)
• Residual fistula
• Late CHF
• Late valvular replacement
• Late death

2
4
36 (1-96) - 3 0 1
29 (1-144) 5 (11) 7 (16) 5 (11) 3 ( 7)
The 3 patients who died w/o surgery had fatal
co-morbid conditions. The remaining 7 patients
did not undergo surgery because they did not have
cardiac failure, severe valvular regurgitation
and echocardiographical abscess.
33
OR 95CI p

• Age gt 65years
• Male gender
• Prosthetic endocarditis
• Symtoms duration gt30 d.
• Moderate or severe CHF
• Renal failure
• Neurologic symptoms
• S.aureus infection
• Vegetation gt10 mm
• Patients with periannular abscess
• Periannular abscess gt 10 mm
• Moderate or severe AR
• Fistulized sinus of Valsalva
• Fistulized cardiac chamber
• EF lt65
• Urgent or emergency surgery

2.8 (1.0-7.9) 0.8 (0.2-2.4) 2.5 (0.9-6.8) 0.8
(0.2-2.6) 2.2 (0.7-5.1) 1.8 (0.7-5.1) 0.6
(0.1-2.8) 1.2 (0.4-3.6) 1.2 (0.4-3.6) 1.6
(0.5-5.5) 2.3 (0.7-7.3) 0.8 (0.3-2.1) - - 1.1
(0.4-3.1) 2.7 (0.9-7.8)
0.05 0.6 0.07 0.7 0.15 0.2 0.5 0.8 0.7 0.4 0.14 0.
7 0.9 0.2 0.8 0.06
34
Limitations
Ascertainment bias multicenter nature
Severity of CHF higher low-grade shunts
underdiagnosed High-risk profles of surgical
candidate Not comparable to medically treated
Not comparing medical and surgical patients
35
Abscesses vs fistulae
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
Kaplan-Meier estimation of survival from time of
diagnosis of periannular complication.
42
Actuarial freedom from death, heart failure
requiring hospital admission and repeat surgery
in patients with periannular complications
surviving the index hospitalization. A. patients
referred to surgical therapy
43
B. patients medically-managed
44
Conclusions

• Aorto-cavitary fistulization in IE is an
  unfrequent event and occurs in patients with
  aortic endocarditis with high grade of local
  tissue destruction.
• It was associated with staphylococci and
  streptococci native-valve IE and with
  coagulase-negative staphylococci prosthetic valve
  IE.
• In-hospital mortality was high even when most
  patients were referred to surgical treatment.
• Congestive heart failure identified the
  subgroup of patients with the worst prognosis.

45

• Prosthetic valve endocarditis
• What?
• When?
• Who?
• Why?

46
(No Transcript)
47
Methods
International Collaboration on Endocarditis
Merged Database Large, multicenter,
international registry of patients with
definite endocarditis by Duke criteria
Clinical, microbiological, echocardiographic
variables to determine Those factors associated
with the use of surgery in PVIE Logistic
regression analysis Propensity score to match
surgery vs medical therapy
48
PVIE Patient characteristics
49
Complications and outcomes of patients with PVIE
50
Propensity analysis of surgical treatment of PVIE
51
Logistic regression analysis of variables
independently associated with in-hospital
mortality in patients with PVIE and matched
propensity for surgical treatment
52
Conclusions
Despite the frequent use of surgery for the
treatment of PVIE this condition continues to be
associated with high in-hospital mortality
After adjustment for factors related to surgical
intervention, brain embolism and S. aureus
infection were independently associated with
in-hospital mortality and a trend toward
a survival benefit of surgery was evident
53
Echocardiographic (TTE) Follow-up
Year Patient TTE TTE FU Last
TTE NYHA Before After (Yrs)
1991 AMG Veg 28 mm Mild TR 13 Severe
TR II Large RV 1991 RPO Veg 22 mm Severe TR
13 Severe TR II Severe TR Large RV Large
RV Large RV 1992 PER Veg 30 mm Severe TR
5 Severe TR I Severe TR Ruptured Large
RV chordae 1994 JLF Veg 22 mm Mild TR
1 Mild TR I 1996 JFG Veg 28 mm Mild TR
1 Severe TR I Severe TR
54
Echocardiographic (TTE) Follow-up
Year Patient TTE TTE FU Last
TTE NYHA Before After (Yrs)
2001 ERA Severe TR Trivial TR pod Po
Death 2002 LML Veg 20 mm Trivial TR
pod Po Death Severe TR Large RV Large
RV 2002 JGR Veg 30 mm Mild TR 2.5 Mild
TR I Severe TR
55
Outcomes
Year Patient FU Drug addiction Recurrent HIV Outco
me (Yrs) relapse endocarditis stage
1991 AMG 6 Yes 14 mos B3 Alive (Corynebacte
rium spp) Late Reop 1991 RPO 6 Yes 48, 58, 63
mos B2 Alive (MSSA all cases) No
Reop 1992 PER 5 No No A2 Alive Late
Reop 1994 JLF 2.5 Yes No A3 Death Overd
ose 1996 JFG 8.5 Yes 7, 12 mos A2 Alive (M
SSA) No reop
56
Year Patient FU Drug addiction Recurrent HIV Outco
me (Yrs) relapse endocarditis stage
2001 ERA PO N N C3 Death 2002 LML PO N
N B2 Death 2002 JGR 2.5 N No A1 Alive

57
The most complex situation
Fibrous Skeletal destruction
58
Acute pectoralis major myositis in an otherwise
healthy young male
59

• 25-year-old male
• Smoker ½ pack/day
• Occasional recreational drugs. NO iv abuse
• Job Waiter. Physically fit. Contact sports
  (judo, full-contact)

• In the past 2 years 4 episodes of abscess
  requiring surgical drainage (hand, foot, knee,
  axilla)
• No other personal nor familiar medical history of
  interest

• 5-day left upper limb and upper left chest pain
  accompanied by high-degree fever (39C), chills
  and malaise

60

• Aortic root replacement with a 20-22 mm
  cryopreserved aortic homograft
• Intraoperative findings Massive AR due to
  perforation of the right coronary cusp on a
  morphologically normal aortic valve. Full root
  subaortic abscess extending towards the left
  atrial roof
• Aortic cross-clamp 73 min CPB 189 min
• Left ventricular failure and myocardial edema
  after CPB. Sternum open. Intraaortic ballon pump
  support

61
Outcome - I

• Postop unstable hemodynamics. Urgent TTE showed
  anterior-septoapical hypokinesia
• Urgent coronary angiogram showed 70 LMCA
  stenosis with remaining normal coronaries
• August 12, 2004 Off-pump LIMA-LAD bypass graft
  and delayed sternal closure
• August 12, 2004 2/2 blood cultures (ORSA)

62
Outcome - II

• Early favourable postop. Improved condition, no
  congestive heart failure
• August 14, 2004, 2/2 negative blood cultures.
  Trasnsferred to ward August 22, 2004. Good
  condition with low-degree fever (37C)
• August 24, 2004 new control TTE

63
Surgery - II

• September 1, 2004 Homograft replacement with a
  21 mm SJM Toronto-Root porcine heterograft
• Surgical findings Subaortic circumferential
  detachment of the normal functioning homograft.
  Extensive lesions of the entire fibrous body.
  Left atrial fistula
• Post-repair severe mitral regurgitation
• Profound left ventricular failure. LVAD Abiomed
  BVS-5000 implanted
• All samples to Microbiology

64
Outcome - IV

• September 2, 2004 Unstable under maximal
  intropic support and LVAD. No further
  conventional surgery indicated. Decision to
  include in emergency WL for heart transplantation
• September 3, 2004 Orthotopic heart
  transplantation

65
Final diagnosis
1. Community-acquired ORSA myositis
2. Acute aortic root ORSA infective endocarditis
3. Heart transplantation
66
Endocarditis and Heart Transplantation

• 1 Galbraith AJ et al. Cardiac transplantation
  for prosthetic valve endocarditis in a previously
  transplanted heart. J Heart Lung Transplant.
  1999 18805-806
• 2 Blanche C et al. Heart transplantation for Q
  fever endocarditis. Ann Thorac Surg. 1994
  581768-1769
• 3 Pulpon LA et al. Recalcitrant endocarditis
  successfully treated by heart transplantation. Am
  Heart J 1994 127958-960
• 4 Park SJ et al. Heart transplantation for
  complicated and recurrent early prosthetic valve
  endocarditis. J Heart Lung Transplant. 1993
  12802-803.
• 5 DiSesa VJ et al. Heart transplantation for
  intractable prosthetic valve endocarditis. J
  Heart Transplant. 1990 9142-143

67
Endocarditis and Heart Transplantation

• Heart transplantation could be an alternative,
  not a contraindication, when in Infective
  Endocarditis all other measures have failed (1)

Galbraith AJ Cardiac transplantation for
prosthetic valve endocarditis in a previously
transplanted heart. J Heart Lung Transplant. 1999
Aug18(8)805-6
68
Case Age Pathogen Valve Position Timing for HTx Conditions
1 25 M. hominis Tissue Aortic 2 months SLE
2 30 S viridans Mechanical Aortic 1 month PreTX cultures
3 58 S viridans Native Mitral 2 years 3 VRs
4 32 C burnetti Native Mi Ao 14 months Persistent fever
5 54 MRSA Mechanical Mitral 17 days Previous HTx
69
Conclusions
IE is a very serious pathology It is not
popular Highly demanding Suboptimal results
Team approach Risk takers
70
Parsonnet score
Single centre Subjective factors
Overestimates risk
Cleveland score
Single centre Excludes non CABG Leads to
gaming
EuroScore
Large multicentre database Fit for all adult
cardiac surgical patients Even correlates with
STS
71
EuroSCORE
Additive
Score mortality
0 2 3 5 6 8 9 11 12gt
0.88 1.51 2.62 3.51 6.51 8.37 14.02
19.12 31.00 42.32
72
EuroSCORE
Its predictive accuracy has been established
Only the additive model has been validated
Inconsistencies among the additive and
logistic models when applied to the high-risk
patients
73
Cross-over point
74
Reasons to predict mortality in Cardiac Surgery
1. Helping to determine indications for
surgery 2. Quality monitoring
Additive EuroScore works well for most purposes
75
Considerations
The relationship between risk factors is not
additive
Combined impact of two or more factors on
operative risk may be more than simple sum
Logistic score more realistic
76
The reality
Infective endocarditis is a high-risk
situation There is lack of data regarding risk
assessment before valve surgery
77
Aim of the study
To validate the EuroSCORE preoperative
stratification risk model in infective
endocarditis
78
Population
Period Jan 95 Jan 04 Patients 147 Mean
age 56.33 15.95 Male gender 69.4
79
Native valve IE
N
Aortic 64 43.5
Mitral 25 17
Tricuspid 2 1.4
Pulmonary 1 0.7
A M 12 8.2
M T 1 0.7
80
Prosthetic valve IE
N
PVE Aortic 17 11.6
Homograft Ao 2 1.4
PVE Mitral 11 7.5
PVE Ao M 1 0.7
PVE Ao PVE Mi 2 1.4
A PVE Mi 1 0.7
81
Intravascular leads
N
DDD 3 2
AICD 1 0.7
VVI R 1 0.7
VVI 2 1.4
Mitral DDD 2 1.4
82
Characteristics
Active endocarditis 91.2 IV Drug
addicts 10.9 HIV 5.4 ESR HD
3.4 Reoperation 27.2
83
Pathogens
N
Culture negative 10 6.8
Staphylococcus 55 37.4
Streptococcus 43 29.3
Enterococcus 14 9.5
Polimicrobial 8 5.4
Candida 1 0.7
Other 14 9.5
84
Type of operation
Emergency 29.9 Urgent 21.8 Elective 46.9

85
EuroSCORE
Additive
Range Mean Median
2 19 10.15 3.81 10
86
EuroSCORE
Logistic
Range Mean Median
1.51 94.17 EM 25.59 20.81 18.95
87
Results
Overall in-hospital mortality 32.7

• Intraoperative death
• 30 days po
• Regardless the length of stay

88
Results
Receiver operating characteristics (ROC) curves
Area SE Lower bound Upper bound Sig.
All patients .826 .036 .756 .896 .000
Asymptotic 95 confidence interval
Area gt 0.7 Good correlation Area gt 0.8 Very
good correlation Area gt 0.9 Excellent
correlation
89
Results
Area SE Lower Bound Upper Bound Sig.
Native valve IE .814 .045 .727 .902 .000
Prosthetic IE .779 .088 .607 .952 .000
90
Results
Area SE Lower Bound Upper Bound Sig.
Aortic position .778 .064 .652 .904 .001
Mitral position .937 .051 .836 1.037 .001
91
Results
Area SE Lower Bound Upper Bound Sig.
Aortic prostheses .729 .125 .484 .980 .112
Mitral prostheses .833 .152 .535 1.132 .068
92
Results
Area SE Lower Bound Upper Bound Sig.
Gram .819 .041 .739 .899 .000
Gram - .833 .204 .433 1.233 .248
93
Results
Area SE Lower Bound Upper Bound Sig.
Staphylococci .834 .054 .727 .940 .000
Streptococci .856 .087 .686 1.026 .002
Enterococci .500 .163 .181 .829 1.000
Polymicrobial .800 .165 .476 1.124 .180
94
Aortic valve
95
Homograft aortic
96
Mitral valve
97
Aortic prosthesis
98
Mitral prosthesis
99
Comments
There is a very good correlation between
logistic EuroSCORE and mortality for the entire
group
Division in subgroups yields a decrease in
statistical power but correlation is almost the
same in all subgroups
The area is good in the prosthetic valve IE
although non significant by position
100
Comments
The area is very good for Gram and
polymicrobial although with low statistical power
There is statistical power for significance in
the Staphylococci and Streptococci groups
101
Limitations
Small sample size
Statistical power decreases when analyzing
subgroups
Just preliminary results
102
When to use Logistic EuroScore?

• To calculate a precise and realistic risk
  prediction for a very high-risk patient,
  particularly when the indication for surgery may
  not be clear

- To monitor quality of care in institutions
where a substantial proportion of patients are of
very high-risk
- To help in the further study of risk modelling
by groups and institutions with a scientific
interest in the subject
- To carry out normal stratification in
institutions with easy availability of accesible
information technology, especially where
high-risk surgery forms a substantial part of the
workload
103
The Future of risk stratification
Larger sample size
More institutions involved
Subgroup analysis (Pathogens, abscess)
Team approach
The role of ICE
Changing our approach to patients?
Quality assurance
104
Conclusions
IE is a very serious pathology It is not
popular Highly demanding Suboptimal results
Team approach Risk takers