Outcome Evidence After
OPCAB Surgery
Overview of Presentation
• The Editors reviewed evidence related to the
following outcomes after CABG surgery
performed on- and off pump:
Early Mortality
Neurocognitive dysfunction
Stroke
Renal failure
Economic comparison between on- and off-pump
CABG
• Prior to this review, the Editors discuss the
Complexity of Generating Evidence and
following the review, they provide their
Conclusions to Date about OPCAB
surgery.
Contents
• Complexity of Generating Evidence
• Early Mortality after CABG
– Studies
– Self-assessment of departmental performance
•
•
•
•
Neurocognitive dysfunction after CABG
Stroke after CABG
Renal failure after CABG
Economic comparison between on- and off-pump
CABG
• Conclusions to date
Coronary Surgery Evolution
• Probably the most frequently performed surgical
procedure worldwide.
• Since its inception (circa 1967), the quality of
anesthesia, anastomosis type and format,
myocardial protection, hospital management,
medical follow-up, and training has continuously
improved.
• Early and late quality of the surgical product can
be represented in mathematical equations (e.g.,
hazard functions) relating to specific outcome
events and intervals.
Challenging Established Practice
• On-Pump: Coronary anastomosis is
performed using the extra-corporeal
circulation to optimize the manipulation of
the heart, the visibility of the coronary
vessels, the stability of the anastomotic area,
the quality of the anastomosis, and the
protection of the heart. Yet, extra-corporeal
circulation itself has potential adverse
effects.
• Off-Pump: The off-pump approach allows
the anastomotic process, but the variability
in approach has created variability in the
quality of the surgical process. Imperfect
technique has forced surgeons to exclude
many patients from the off-pump approach
and might have compromised the
anastomotic quality. Yet, outcome when
performed by off-pump experts is promising
Need for Standardization
• Just as occurred with on-pump coronary surgery,
the generation of improved outcome after an offpump approach in coronary surgery needs to be
preceded by a standardization of the off-pump
technique.
• This will allow complete manipulation of the heart,
perfect visibility of the coronary vessels, stability
of the anastomotic area, and protection of the heart.
Need for Standardization
• Only then can a comparable anastomotic
quality be achieved with possible avoidance
of some of the deleterious aspects of the
extra-corporeal circulation.
Issue 1: Is there a fixed risk for
CABG?
• There is a wide spectrum of risk
– Low risk for large patient cohorts
– Increasing risk for cohort subset, most frequently due to
increased co-morbidity and severity of disease
• Outcome varies with risk
Risk of the
procedure
General
Cohort
Distribution of risk
Increasingly
Complex
Subgroup
Issue 2: How to score risk?
• Risks have been structured in “languages of risk”
– (Example: STS risk, EuroSCORE, Parsonnet, etc.)
• These risk languages can be used to systematically
compare on- and off-pump risk
Standard Risk Languages
(STS, EuroSCORE, Parsonnet)
On-pump
approach
Off-pump
approach
Issue 3: How certain is an
observation?
• Every observation is associated with a degree of
uncertainty around whether that observation in the
sample population represents the true value in the
total population.
• In the scientific literature, we usually accept a 5%
level of uncertainty. This is expressed as a 95%
Confidence Limit or Interval around the
observation.
Issue 3: How certain is an
observation?
• The degree of uncertainty, or size of the Confidence
Interval, is inversely related to the total sample size
of the population in which the observation was
made.
• For example, if you derive a value of 0 from a
sample of 20 subjects, you are 95% confident that
the real value lies between –17 and 17. In a sample
of size of 200, you are 95% confident that the value
lies between –2 and 2.
Issue 4: How long should we
observe?
• The interval for a peri-procedural risk
should include the interval until the early
hazard function stabilizes into a constant
risk, independent of hospital discharge.
Minimum Study Intervals Following CABG
Outcome
Early Mortality
Stroke
Infarction
Atrial fibrillation
Study Interval
3 months
8 days
8 days
6 days
Issue 4: How long should we
observe?
• To balance “early costs” with “late benefits”, the
ideal study interval is 5-10 years. The most costly
procedure may be the most beneficial.
H o sp ita l M o rta lity
0 ,0 2
3 0 -D a y M o rta lity
0,0 8
Ea r ly P e ri-p ro c e d u ra l R isk
0 ,3
Q u a lity o f C a r e
10
0
2
4
6
Y e ar s
8
10
12
Issue 5: How large a sample size?
• Sample size is function of the
–
–
–
–
α type I error allowed
β type II error allowed
actual predicted risk
expected reduction of risk
Predicted Risk
1%
2% 3% 4%
10%
10% risk
reduction
197,7 97,9 64,6 48,0
50
24 49 11
18,06
4
50% risk
reduction
6,253 3,10 2,04 1,52
0
9
4
578
Issue 5: How large a sample size?
• For example, what is the estimated sample size of
each arm of a clinical trial, if the tolerated α type I
error is 0.05 and β type II error is 0.1?
10% risk
reduction
Predicted Risk
1% 2% 3% 4%
197,7 97,9 64,6 48,0
50
24 49 11
50% risk
reduction
6,253 3,10 2,04 1,52
0
9
4
10%
18,06
4
578
Issue 5: How large a sample size?
10% risk
reduction
Predicted Risk
1% 2% 3% 4%
197,7 97,9 64,6 48,0
50
24 49 11
50% risk
reduction
6,253 3,10 2,04 1,52
0
9
4
10%
18,06
4
578
Early Mortality After CABG:
Studies
• Reference: Elimination of cardiopulmonary
bypass improves early survival for multivessel coronary artery bypass patients.
Magee M.J., Jablonski K.A., Stamou S.C.,
et al. Ann Thorac Surg 2002;73:1196-1203.
Early Mortality After CABG:
Studies
• Message: A mathematical model, for hospital
mortality after CABG, confirmed several known
risk factors. In addition, cardio-pulmonary bypass
was independently associated with an increased
risk for mortality, with an odds ratio of 1.79 (95%
CL 1.2-2.7).
• Of Interest: Multi-center (N=2), large cohort of
OPCAB (N=1983), multi-vessel disease,
multivariate logistic regression analysis
Early Mortality After CABG:
Studies
• Limitations: Biased selection process of
patients (somewhat corrected by propensity
scoring), short and biased observation
interval (hospital stay)
Early Mortality After CABG:
Studies
• Reference: Off-pump coronary artery bypass is associated
with improved risk-adjusted outcomes. Plomondon ME,
Cleveland JC, Ludwig ST, et al. Ann Thorac Surg.
2001;72:114-119.
• Message: Centers report a 39% reduction of the STS-risk
predicted 30-day mortality in a selected off-pump
population. A reduction of only 10% of the STS-predicted
30-day mortality was identified in the on-pump population.
• Of Interest: Multi-center (N=9), reasonable cohort of
OPCAB (N=680), 66% of OPCAB patients had 3-vessel
disease, multivariate logistic regression analysis
Early Mortality After CABG:
Studies
• Limitations: Biased selection process of patients
(somewhat corrected for by an indicator variable),
short and biased observation interval (30-day)
Early Mortality After CABG:
Studies
• Reference: Off-pump coronary artery bypass grafting
decreases risk-adjusted mortality and morbidity. Cleveland
J.C., Shroyer L.W., Chen A.Y., et al. Ann Thorac Surg
2001;72:1282-1289.
• Message: Patients, after STS-risk adjustment, receiving
off-pump procedures were less likely to die (adjusted oddsratio 0.81, 95% CL 0.70-0.91).
• Of Interest: Multi-center (N=126), large cohort of
OPCAB (N=11,717), STS-risk adjustment
Early Mortality After CABG:
Studies
• Limitations: Biased selection process of patients,
mostly single and two-vessel disease OPCAB
patients, short and biased observation interval (30day)
Early Mortality After CABG:
Studies
• Reference: Outcomes experience with offpump coronary artery bypass surgery in
women. Brown P.P., Mack M.J., Simon,
A.W., et al. Ann Thorac Surg
2002;74:2113-2119.
Early Mortality After CABG:
Studies
• Message: After controlling for patient,
procedure, medication, time period and site
characteristics, the estimated odds ratio
indicates that women undergoing on-pump
CABG surgery experienced a 42% higher
mortality rate than women undergoing
OPCAB (P=0.023).
Early Mortality After CABG:
Studies
• Of Interest: Focused on the female
population, multi-center (N=78), large
cohort of OPCAB (N=2631), multivariate
logistic regression risk-adjustment
Early Mortality After CABG:
Studies
• Limitations: Biased selection process of
patients, short and biased observation
interval (hospital stay)
Early Mortality After CABG:
Studies
• Reference: Safety and efficacy of off-pump
coronary artery bypass grafting. Arom K.V.,
Flavin T.F., Emery R.W., et al. Ann Thorac
Surg 2000;69:704-710.
Early Mortality After CABG:
Studies
• Message: No difference in operative mortality is
identified after STS-risk predicted grouping in the
low (0-2.6% predicted risk, N=216) and medium
(2.6-10% predicted risk, N=95) risk group. A
significant difference is observed in the high (1020% predicted risk, N=39) risk group, with a 73%
reduction of the risk versus on-pump.
• Of Interest: STS-risk grouping in three categories.
Early Mortality After CABG:
Studies
• Limitations: One center, limited cohort of
patients (N=350), biased selection process of
patients, only 1.6 anastomoses per patient in high
risk OPCAB patients, short and biased observation
interval (hospital stay). The late follow-up section
at 1 year has only a 66% completeness of followup.
Early Mortality After CABG:
Self-Assessment
VLAD or CRAM Plot - Reference
• Reference: Monitoring the results of
cardiac surgery by variable life-adjusted
display. Lovegrove J., Valencia O., Treasure
T., et al. Lancet 1997; 350(9085):1128-1130.
Early Mortality After CABG:
Self-Assessment
• Message: VLAD provides a graphical display of
risk-adjusted survival figures for individual
surgeons or units over time and could be modified
to monitor performance over a range of treatments
and outcomes.
• Of Interest: Easy calculation, nicely readable,
completed within a simple spreadsheet
Early Mortality After CABG:
Self-Assessment
• Limitations: Minimum number of patients
required, levels of uncertainty not
calculated, limited to the quality and
interval of the scoring system, does not give
information of performance across the
spectrum of risk
Early Mortality After CABG:
Self-Assessment
VLAD or CRAM Plot - Calculation
 Calculate the individual patient’s risk, using any scoring
system.
 Express this in lives: 5% predicted risk is 0.05 “Lives”.
 If the patient survives: recalculate value as + 0.05 “Lives”
 If the patient dies in hospital: recalculate value as 1-risk = 10.05 = -0.95 “Lives”
Early Mortality After CABG:
Self-Assessment
 Cumulate the obtained values
 In the example patient 6 dies
 Patient 6 has a 10%
predicted risk.
Creation of a Variable Life-adjusted Display
Lives Saved > Scoring System
0,4
0,21
0,2
0,12
0
0,13
0,13
0,05
Pt 1
Pt 2
Pt 3
Pt 4
Pt 5
Pt 6
-0,2
-0,4
-0,6
Individual Risk
-0,71
-0,8
Early Mortality After CABG:
Self-Assessment
VLAD or CRAM Plot - Example
L iv e s s a v e d v e rs u s s c o rin g s y s te m
30
25
20
15
10
5
o ff-p u m p
o n -p u m p
0
1
-5
201
401
601
801
1001
1201
1401
Early Mortality After CABG:
Self-Assessment
• Reference: The challenge of departmental quality
control in the reengineering towards off-pump
coronary artery bypass grafting. Sergeant P., de
Worm E., Meyns B., et al. Eur J Cardio-thorac
Surg 2001;20:538-543.
• Message: The graphical depictions provide a
cumulative insight in the performance across the
spectrum of risk.
Early Mortality After CABG:
Self-Assessment
• Of Interest: Easy calculation, nicely readable,
completed within a simple spreadsheet
• Limitations: Provides cumulative insight across
risk sectors, but not within individual risk sectors,
due to large uncertainties related to low N in high
risk categories. Limited to the quality and interval
of the scoring system.
Early Mortality After CABG:
Self-Assessment
Performance Across the Spectrum of Risk Example
• The cumulative deviation, reduction or increase,
of the observed versus the predicted hospital
mortality is calculated in a stepwise manner. At
each step a 1% higher predicted risk category is
added to the population. At the extreme right of
the plot the total population is included.
Early Mortality After CABG:
Self-Assessment
Performance Across the Spectrum of Risk Example
• Plot the cumulative observed mortality in %
versus the cumulative predicted mortality in %.
• Start with the patients, having a predicted risk
between 0 and 1, calculate their observed %
mortality and plot versus the predicted mortality.
Early Mortality After CABG:
Self-Assessment
• Add the patients with a
risk between 1 and 2%
and plot again.
• Plot again at each
step of one % of
predicted risk.
4
o n -p u m p
o ff-p u m p
3 ,5
% o b s e rv e d m o rt a lit y
3
2 ,5
2
1 ,5
1
0 ,5
0
0
0 ,5
1
1 ,5
2
2 ,5
3
3 ,5
% p r e d ic te d m o r ta lity
4
4 ,5
5
5 ,5
Neurocognitive Dysfunction
After CABG
• Reference: Assessment of neurocognitive
impairment after off-pump and on-pump
techniques for coronary artery bypass graft
surgery: prospective randomized controlled
trial. Zamvar V., Williams D., Hall J. et al.
BMJ 2002;325:1268-1273.
Neurocognitive Dysfunction
After CABG
• Message: Patients were considered to have
neurocognitive impairment if they showed a
deterioration of 1 SD or more in two or more tests.
One week postop, 27% in the off-pump and 66%
in the on-pump had neurocognitive impairment
(P=0.004). Ten weeks postop, 10% of the offpump and 40% of the on-pump had
neurocognitive impairment (P=0.017).
Neurocognitive Dysfunction
After CABG
• Of Interest: Randomized trial, limited to
triple vessel disease patients, nine standard
tests.
• Limitations: Limited to the first 10 weeks
Neurocognitive Dysfunction
After CABG
• Reference: Neuromonitoring and
neurocognitive outcome in off-pump versus
conventional coronary bypass operation.
Diegeler A, Hirsch R., Schneider F., et al.
Ann Thorac Surg 2000;69:1162-1166.
Neurocognitive Dysfunction
After CABG
• Message: Postoperative CSS scoring was not
different between on- and off-pump groups
(P=0.2). Psychiatric assessment scoring between
on- and off-pump group was significant (P=0.04).
Syndrom Kurtz Test (cognition) scoring was
highly significant (P=0.0001) in favor of offpump. The median number of HITS in the onpump group was 394 versus 11 in the off-pump
(P<0.0001).
Neurocognitive Dysfunction
After CABG
• Of Interest: Randomized trial, 3 standard
neurocognitive tests, investigation using
high intensive transient signal processing
(HITS)
• Limitations: Limited to the day 1 and 7
after surgery
Neurocognitive Dysfunction
After CABG
Early Dysfunction
• Reference: Cognitive outcome after off-pump and onpump coronary artery bypass graft surgery, a randomized
trial. Van Dijk D., Jansen EW., Hijman R., et al. JAMA
2002;287:1405-1412.
• Message: At 3 months after surgery, cognitive decline
occurred in 21% of the on-pump and 29% of the off-pump
patients (P=.15). At 12 months, cognitive decline occurred
in 31% of the off-pump patients and 34% after on-pump
CABG.
Neurocognitive Dysfunction
After CABG
• Of Interest: Randomized trial, 11 standard
neurocognitive tests, testing at 3 and 12 months
• Limitations: Largely patients with single and
two-vessel disease and low risk for cerebral
dysfunction
Neurocognitive Dysfunction
After CABG
Early Dysfunction
• Reference: Serum S-100 protein release and
neuropsychologic outcome during coronary
revascularization on the beating heart: a prospective
randomized study. Lloyd C.T., Ascione R., Underwood
M.J., et al. J Thorac Cardiovasc Surg 2000;119:148154.:148-154
• Message: There were no significant differences between
on-pump and off-pump groups in the magnitude of change
across all 7 dimensions of neurocognitive outcome at 12
weeks (P=.18).
Neurocognitive Dysfunction
After CABG
• Of Interest: Randomized trial, 7 standard
neurocognitive tests, testing at 12 weeks after
surgery
• Limitations: Patients without known neurological
abnormality
Stroke After CABG
Early Stroke
• Reference: Stroke after conventional versus
minimally invasive coronary artery bypass.
Stamou S.C., Jablonski K.A., Pfister A.J. Ann
Thorac Surg 2002;74:394-399.
• Message: After adjustment for preoperative riskvariability through propensity score matching, an
odds-ratio for stroke of 1.8 is identified in the onpump versus the off-pump group.
Stroke After CABG
• Of Interest: Propensity matching possible
for 72% of the OPCAB patients, OPCAB
N=1670 in each group, logistic regression
analysis, interesting and very complete
model building
• Limitations:
Stroke After CABG
Early Stroke
• Reference: Off-pump coronary artery bypass
grafting decreases risk-adjusted mortality and
morbidity. Cleveland J.C., Shroyer L.W., Chen
A.Y., et al. Ann Thorac Surg 2001;72:1282-1289.
• Message: Patients, with known cerebrovascular
disease, receiving off-pump procedures were less
likely to have a stroke (4.6% with conventional
CABG and 2.5% in the off-pump group).
Stroke After CABG
• Of Interest: Multi-center (N=126), large cohort of
OPCAB (N=1523 with known cerebrovascular
disease), acceptable observation interval (30-day).
• Limitations: Biased selection process of patients,
mostly single and two-vessel disease OPCAB
patients, no additional risk-adjustment beyond
cohort with CVD disease.
Stroke After CABG
Early StrokeMessage: Cardiopulmonary
bypass was an independent risk factor for
focal neurologic deficit, with an odds ratio
of 3.82 (95% confidence interval, 1.4 to
10.3; p=0.005). Aortic manipulation did not
significantly influence neurologic outcome
in off-pump patients.
Stroke After CABG
• Of Interest: Multi-center (N=2), large cohort of
OPCAB (N=1117), acceptable observation
interval (hospital stay), multivariable logistic
regression for risk-adjustment, adjustment in
aortic manipulation.
• Limitations: Biased selection process of patients
(corrected for by using propensity scoring)
Stroke After CABG
Early Stroke
• Reference: Safety and efficacy of off-pump
coronary artery bypass grafting. Arom K.V.,
Flavin T.F., Emery R.W., et al. Ann Thorac
Surg 2000;69:704-710.
• Message: Several postoperative events are
studied.
Stroke After CABG
• Of Interest: There were no significant differences
in the number of patients who suffered from
neurological deficits such as permanent stroke (2.0
% on-pump versus 1.4 % off-pump, p=0.42) and
transient ischemic attack (0.9 % on-pump versus
0.3 % off-pump, p=0.35).
• Limitations: One center, limited cohort of
patients (N=350), biased selection process of
patients, no correction for stroke-risk variability in
populations
Renal Failure After CABG
• Reference: Does off-pump coronary
surgery reduce morbidity and mortality?
Sabik J.F., Gillinov A.M., Blackstone E.H.,
et al. J Thorac Cardiovasc Surg
2002;124:698-707.
Renal Failure After CABG
• Message: Postoperative mortality, stroke,
myocardial infarction and reoperation for
bleeding was similar in on- versus off-pump
patients. There was significantly more
encephalopathy (P=0.02), sternal wound
infection (P=0.04), red blood cell use
(P=0.002) and renal failure requiring
dialysis (P=0.03) in the on-pump patients.
Renal Failure After CABG
• Of Interest: Propensity-matched pairing of
datasets, reasonable N OPCAB (406).
• Limitations: One center, selected patients
in original datasets, fewer anastomoses in
the off-pump population, even after
matching.
Renal Failure After CABG
• Reference: On-pump versus off-pump
coronary revascularization: evaluation of
renal function. Ascione R., Lloyd C.T.,
Underwood M.J., et al. Ann Thorac Surg
1999;68:493-498.
Renal Failure After CABG
• Message: The creatinine clearance decreased
more (P=0.0004) in the first postoperative 48
hours in the on-pump group. The urinary NAG
(N-acetyl-β-glucosaminidase) activity values
remained significantly higher (P=0.0272) and the
albumin-to-creatinine ratio was worse (P=0.0083),
in the postoperative 24-48 hours, in the on-pump
versus the off-pump population.
Renal Failure After CABG
• Of Interest: Prospective randomized trial,
refined analysis of renal function
• Limitations: One center, population at low
risk for dialysis or severe renal failure.
Renal Failure After CABG
• Reference: Safety and efficacy of off-pump
coronary artery bypass grafting. Arom K.V.,
Flavin T.F., Emery R.W., et al. Ann Thorac
Surg 2000;69:704-710.
Renal Failure After CABG
• Message: The incidence of new renal failure was
greater in on-pump patients overall, and in each
risk group, but without any statistical significance,
except in the high risk patients.
low risk 4% on-pump 3% off-pump (P=0.49)
medium risk
10% on-pump
10% off-pump
(P=0.79)
high risk 21% on-pump
3% off-pump (P=0.006)
Renal Failure After CABG
• Of Interest: STS-risk grouping in three categories
• Limitations: One center, limited cohort of
patients (N=350) certainly in high risk, biased
selection process of patients, only 1.6 anastomoses
per patient in high risk OPCAB patients
Economic Comparison Between
On- & Off-Pump CABG
• Reference: Economic outcome of off-pump
coronary artery bypass surgery: a
prospective randomized study. Ascione R.A.,
Lloyd C.T., Underwood M.J., et al. Ann
Thorac Surg 1999;68:2237-2242.
Economic Comparison Between
On- & Off-Pump CABG
• Message: Operative costs were significantly lower
in the off-pump group. Bed occupancy and
nursing costs account for the largest saving in the
off-pump group. Blood loss and transfusion
requirements were significantly less in the offpump group. The costs for the management of
postoperative complications were significantly
higher in the on-pump group.
Economic Comparison Between
On- & Off-Pump CABG
• Of Interest: Prospective randomized trial,
detailed cost calculation
• Limitations: One center, limited cohort of
patients in each arm (N=100), limited
number of complications
Economic Comparison Between
On- & Off-Pump CABG
• Message: Cardiopulmonary bypass was
found, in multiple regression models, to be
an independent predictor of both increased
postoperative length of stay (p<0.0001) and
increased hospital stay (P=0.0048).
Economic Comparison Between
On- & Off-Pump CABG
• Of Interest: Consecutive series compared with a
matched control group of 1000 patients, economic
outcome linked to clinical data, multiple
regression model for risk-adjustment
• Limitations: One center, limited number of
complications
Economic Comparison Between On& Off-Pump CABG
• Reference: Off-pump surgery decreases
postoperative complications and resource
utilization in the elderly. Boyd W.D., Desai N.D.,
Del Rizzo D.F., et al. Ann Thorac Surg
1999;68:1490-1493.
• Message: A savings of 14% of the hospital cost
was obtained in the off-pump approach. The cost
of the procedure was based on the fixed operating
room and supply costs plus the variable hospital
and ICU bed costs. Professional fees were not
included.
Economic Comparison Between On& Off-Pump CABG
• Of Interest: Consecutive series of elderly (age >
70 years), economic outcome linked to clinical
data
• Limitations: One center, small series (N off-pump
= 30, N on-pump = 60), no actual risk adjustment
but both groups were found to be comparable
before surgery
OPCAB Surgery: Conclusions to
Date
• Coronary surgery off-pump is performed in
hundreds of centers across the world. The
evidence is structured in > 1000 peerreviewed articles.
OPCAB Surgery: Conclusions to
Date
• The selection of the patients is related to the
experience of the center and to the degree of
process re-engineering that has taken place.
• Coronary surgery off-pump reduces, after
adjustment for variability in risk, early mortality
and some major morbidity events: neurocognitive
dysfunction, stroke and renal failure.
OPCAB Surgery: Conclusions to
Date
• Off-pump coronary surgery allows and
mandates a rigorous re-engineering of the
surgical production process. Only then can
there also be an improvement of the
economic performance.
• Coronary surgery off-pump is here to stay.
Descargar

Outcome Evidence After OPCAB Surgery January 2003