Professional Documents
Culture Documents
Field MJ, Lohr KN (eds). Clinical Practice Guidelines: Directions for a New
Program. Institute of Medicine, Washington, DC: National Academy Press,
1990.
Guidelines and
Performance Measures
• Public reporting
– Cardiac surgery outcomes in New York State
• Pay for performance
– Reward “good” behavior
– CMS: several VTE prevention P4P measures
• Registries
– Accreditation
– Facilitate quality improvement
Case Scenario
• 50 year-old woman scheduled to undergo elective
laparoscopic cholecystectomy
– PMH notable for moderate emphysema
– No personal or family history of VTE
– Medications: Spiriva®, albuterol
– Stopped smoking 1 year ago
• What should we recommend for perioperative
VTE prophylaxis in this patient?
BMJ. 1999;318:593-596.
Conceptual Framework
• VTE pharmacoprophylaxis involves a tradeoff
between preventing thrombosis and causing bleeding
• When making tradeoffs, need to compare absolute
risks of thrombosis and bleeding
• In order to determine absolute risks (eg, number of
symptomatic DVTs prevented), need to know the
following:
– Baseline risk in control/comparison group
– Relative risk for intervention vs control
• When making tradeoffs, also need to assign values to
events being compared
Calculating Absolute Effects
Scenario Baseline Baseline RR VTE RR Bleed Number Number
Risk of Risk of (UFH vs (UFH vs of VTEs of Bleeds
sVTE Major no no Prevented Caused
Bleeding Prophy) Prophy) (per 1000) (per 1000)
(%) (%)
Any surgical
? ? 0.50 2.0 ? ?
patient
Calculating Absolute Effects
Scenario Baseline Baseline RR VTE RR Bleed Number of Number of
Risk of Risk of (UFH vs (UFH vs VTEs Bleeds
sVTE Major no no Prevented Caused
Bleeding Prophy) Prophy) (per 1000) (per 1000)
(%) (%)
Moderate VTE/ 2 1 10 10
Average Bleed
Moderate VTE/ 2 2 10 20
High bleed
0.50 2.0
High VTE/ 4 1 20 10
average bleed
High VTE/ 4 2 20 20
high bleed
PICO Question
• Among patients undergoing elective abdominal
surgery, should LDUH vs no prophylaxis be used
for VTE prevention?
• Are we confident that the benefits of reducing
fatal and nonfatal VTE exceed the harms of
increasing fatal and nonfatal major bleeding?
Evidence Synthesis
• Systematic Review for each PICO!
– Literature search
– Assessment for eligibility
– Assessment of study quality
– Data abstraction and synthesis
• Expensive, time-consuming, and labor-intensive
– AHRQ Evidence-Based Practice Centers
Assessment of Baseline Risk
Ideal study Most studies
• Large, population-based • Single center
• Prospective (?) • Referral center bias
• Few exclusions or losses to • Prospective
follow-up • Employ surveillance
• Well-defined endpoints methods to identify
– Important to patients asymptomatic DVT
– Objectively confirmed • Short time horizon
• Sufficiently long time horizon • No description or
• No prophylaxis or controlled for adjustment for
prophylaxis prophylaxis
Estimating Baseline Risk
4
Baseline Risk of VTE
52%
• Frequency
– Q8h 72 +/- 5
– Q12h 63 +/-5
Calculating Absolute Effects
Scenario Baseline Baseline RR VTE RR Number Number
Risk of Risk of (UFH vs Bleed of VTEs of Bleeds
sVTE Major no (UFH vs Prevented Caused
Bleeding Prophy) no (per 1000) (per 1000)
(%) (%) Prophy)
Moderate
VTE-Risk 2.6 1.2 0.59 1.57 11 7
Patient
Evidence Synthesis:
Fatal bleeding
7 randomised no serious no serious no serious serious Moderate
trials limitations inconsistency indirectness
Nonfatal major bleeding, inferred from excessive intraoperative bleeding or need for transfusion
BMJ. 2008;336:1049-1051.
Recommendation
For general surgical patients at moderate risk for
venous thromboembolism who are not at high risk
for perioperative bleeding, we suggest low-dose
unfractionated heparin (Grade 2B) over no
prophylaxis.
Recommendation
For general surgical patients at high risk for
venous thromboembolism who are not at high risk
for perioperative bleeding, we recommend use of
low-dose unfractionated heparin (Grade 1B) over
no prophylaxis.
Major Challenges
• Multiple sources of heterogeneity
• Indirectness
– When should one apply indirect evidence from studies
performed in a mixed (surgical) or different patient
population?
• Surrogate outcome: asymptomatic DVT
• Poorly standardized outcome: major bleeding
• Limited information about baseline risk of VTE in
absence of prophylaxis
Biases Introduced by Surveillance for
Asymptomatic DVT
• Downward: identification and treatment of
asymptomatic DVT prevents unknown number of
events that would have become symptomatic
• Upward: more likely to label a finding (eg, leg
swelling) as a symptomatic if event is detected by
surveillance
• Difficult to estimate the ratio of asymptomatic to
symptomatic events
Additional Challenges
• Numerous comparisons
– LDUH, LMWH, fondaparinux, low-dose ASA, high-dose
ASA, ES and IPC vs no prophy
– Mechanical vs pharmacologic
– Add mechanical to pharmacologic
– 16 unique evidence profiles and still counting!
• Numerous surgical populations
– Abdomen and pelvis (vascular, bariatric)
– Neurosurgery (craniotomy, spine)
– Trauma (TBI, SCI, other major trauma)
– Cardiac, thoracic, other…
Patients Undergoing General, GI, Urologic, Gynecologic, Bariatric, Vascular,
Plastic, or Reconstructive Surgery
For thoracic surgery patients at high risk for VTE who are not
at high risk for perioperative bleeding, we suggest LDUH
(Grade 1B) or LMWH (Grade 1B) over no prophylaxis. In
addition, we suggest that mechanical prophylaxis with ES or
IPC should be added to pharmacologic prophylaxis (Grade 2C).
Patients Undergoing Thoracic Surgery
For thoracic surgery patients who are at high risk for major
bleeding, we suggest use of mechanical prophylaxis, preferably
with optimally applied IPC, over no prophylaxis until the risk
of bleeding diminishes and pharmacologic prophylaxis may be
initiated (Grade 2C).
Patients Undergoing Craniotomy
For craniotomy patients at very high risk for VTE (eg, those
undergoing craniotomy for malignant disease), we suggest
adding pharmacologic prophylaxis to mechanical prophylaxis
once adequate hemostasis is established and the risk of bleeding
decreases (Grade 2C).
Patients Undergoing Spinal Surgery