Anticoagulation Update - University of Chicago

Download Report

Transcript Anticoagulation Update - University of Chicago

The Program in
Hospital Medicine
Anticoagulation Update
David Lovinger, MD
Assistant Professor
Program in Hospital Medicine
University of Chicago
I have no financial conflicts to disclose
Objectives

To learn about new developments in
anticoagulation therapy, monitoring and safety.
• National Patient Safety Goal 3E
• Pharmacogenetic dosing of warfarin
• Use of very-low dose vitamin K to stabilize INR in
hard to control patients
• Duration of anticoagulation for patients with VTE
Introduction



In both its use and avoidance,
anticoagulation is a risky therapy.
Warfarin is a very hard drug to dose
properly.
There are many opportunities to improve
the quality of care.
Risk of Adverse Events for
Anticoagulated Patients
Adapted from Hylek EM et. al. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients
with nonrheumatic atrial fibrillation. NEJM 1996; 335(8):540-6 and Hylek EM and Singer DE. Risk factors for
intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994; 120(11):897-902.
Anticoagulation Clinics

JCAHO Patient Safety Goal 3E:
• Mandate:“Reduce the likelihood of patient harm associated
with the use of anticoagulation therapy.”
• Rationale: Anticoagulation is a high risk treatment, which
commonly leads to adverse drug events due to the
complexity of dosing these medications, monitoring their
effects, and ensuring patient compliance with outpatient
therapy. The use of standardized practices that include
patient involvement can reduce the risk of adverse drug
events associated with the use of heparin (unfractionated),
low molecular weight heparin (LMWH), warfarin, and other
anticoagulants
Key Ingredients to NPSG




“a defined anticoagulation management program”
“approved protocols”
Proper monitoring (baseline and follow-up INRs)
Education of patients and families about
• Dietary restrictions
• Monitoring/follow-up
• Drug interactions


Outpatient and inpatient components*
Evaluation and monitoring of the program
* Not explicitly part of the NPSG, but UCMC has interpreted it to include an
inpatient component.
Timetable
1.
2.
3.
4.
April 1, 2008: The organization’s leadership has
assigned responsibility for oversight and coordination
of the development, testing, and implementation of
NPSG Requirement 3E.
July 1, 2008: An implementation work plan is in place
that identifies adequate resources, assigned
accountabilities, and a time line for full
implementation by January 1, 2009.
October 1, 2008: Pilot testing in at least one clinical
unit is under way.
January 1, 2009: The process is fully implemented
across the organization.
Do Clinics Work?

In general, there is an increase in time spent in
therapeutic range:
• Time in therapeutic range – standard: 50-55%
• Time in therapeutic range – clinic: 60-65%
• Reduction in time spent in “alert” range (INR > 4.0 or <
1.4).1


Even self-testing is better than usual care.2
In studies of pharmacist managed AC clinic:
• 39% fewer anticoagulation-related complications.3
• Pt INR’s were in therapeutic range 20% more often. 3
• Reduction in hospitalizations.4
1
2
3
4
Personal communication, University of Utah, Chronic Anticoagulation Clinic.
Connock M, et al, Health Technol Assess, 2007.
Witt, et al, Chest, 2005.
Locke, et al, Pharmacotherapy, 2005.
Do Clinics Work?

Yes, but:
• The model is important: pharmacists are better
studied than RN’s.
• Standardization and continual evaluation are
essential.
• Centralization is likely NOT essential.
• Small improvements in INR result in significant
clinical outcomes.
Introduction to Warfarin
Pharmacogenetics

Background on common genetic variants and
how they affect warfarin dosing.
• VKORC1 – affects vitamin K metabolism
• CYP2C9 – affects warfarin metabolism


How these apply to clinical decisions.
Recent evidence regarding the use of warfarin
dosing algorithms.
Warfarin Pharmacology

CYP2C9---|
Hall, 2006
2 main components to the
action of warfarin in the
body:
• Warfarin inhibits Vitamin
K Epoxide Reductase
(VKOR) which helps
recycle Vit K.
• Warfarin is metabolized via
the cytochrome P450
system, specifically,
CYP2C9.
CYP2C9 Variant Alleles
Impact Warfarin Metabolism



CYP2C9*2 has
30-50% activity of
wt (CYO2C9*1.)
CYP2C9*3 has
10% activity of wt.
These variants
account for 12% of
the variation in
warfarin dose.
Dervieux, 2005
CYP2C9 Allele Frequencies
Vary Between Populations




Population
CYP2C9*1 (WT)
CYP2C9*2
CYP2C9*3
Caucasian
78 – 84%
10 - 12.5%
5 – 10%
Asian
97.8%
-
2.2%
African-American
93.0%
5.6%
1.4%
CYP2C9*2 reduces the rate of metabolism resulting in lower
clearance (30-50% activity of WT).
CYP2C9*3 has more significantly reduced rate of metabolism
and decreased enzyme efficiency (10% activity of WT).
Both result in lower need for warfarin and a lower dose
Non-WT patients have an increased risk of supratherapeutic
INRs and bleeding (Higashi, et al, JAMA, 2002).
VKORC1 Variants Have Functional
Consequences on Warfarin Dose

VKORC1 haplotype
alone accounts for 21-25%
of the variability in
warfarin dose.
• Hap A/A = 2.7±0.2 mg/day
• Hap A/B = 4.9±0.2 mg/day
• Hap B/B = 6.2±0.3 mg/day


Mechanism of action in
these variants is unclear.
Variant VKORC1
haplotypes do not appear
to have the same bleeding
risk of CYP2C9 variants.
Reider, 2005
Frequencies of VKORC1
Variants



Form two distinct
groups.
Hap A is
associated with a
lower warfarin
maintenance dose.
Hap B is
associated with a
higher warfarin
maintenance dose.
Population
Hap A
Hap B
European
37 - 42%
57 - 58%
African
American
14 - 21%
49 - 58%
Asian
85 - 89%
10 - 14%
Peruvian
27%
71%
Mexican
38%
57%
African
23%
49%
Variables Known to Influence
Warfarin Dose
Marsh (2006)
Algorithm for Dosing

Daily dose of warfarin=
exp[(0.9751-0.3238 x VKOR3673G>A ) + (0.4317 x BSA) –
(0.4008 x CYP2C9*3) – (0.00745 x age) – (0.2066 x
CYP2C9*2) + (0.2029 x target INR) – (0.2538 x
amiodarone) + (0.0922 x smokes) – (0.0901 x AA race) +
(0.0664 x VTE)]


Difficult to use in clinical practice
Easier to use in clinical practice:
http://www.warfarindosing.org/

Does it work?
The Program in
Hospital Medicine
Randomized Trial of GenotypeGuided Versus Standard Warfarin
Dosing in Patients Initiating Oral
Anticoagulation
Anderson JL, Horne BD, Stevens SM, et al.
Circulation, 2007;116:2563-2570
Objective

To compare the effect of genotype-guided
dosing on INR to standard, nomogram-based
dosing.
Methods

206 inpatients in academic medical center.
• Inclusions: patients with indication for AC (Goal INR 2-3)
• Exclusions: rifampin, advanced age, renal or hepatic disease
• Blinding: Unblinded



Randomized to PG-dosing or standard care based on
10 mg nomogram.
Primary endpoint was percentage of INR’s out of
range.
Secondary endpoints include time to supratherapeutic
INR, time in range, percentage of patients at goal by
days 5 and 8, number of dosing changes.
Risk of Adverse Events for
Anticoagulated Patients
Adapted from Hylek EM et. al. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients
with nonrheumatic atrial fibrillation. NEJM 1996; 335(8):540-6 and Hylek EM and Singer DE. Risk factors for
intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med 1994; 120(11):897-902.
Results


With the exception of the
number of dosing changes,
there was no difference
between the 2 groups.
A non-significant trend
towards fewer INR’s and
fewer supratherapeutic
INR’s, but...
• Small study
• Trends are potentially
significant

Bottom line: not ready for
widespread use, but will
likely be useful in the future.
The Program in
Hospital Medicine
Genetic Determinants of
Response to Warfarin During
Initial Anticoagulation
Schwarz UI, Ritchie MD, Bradford Y, et al.
N Engl J Med 2008;358:999-1008.
Objective

To compare the effect of genotype-guided
dosing on INR and bleeding to standard,
nomogram-based dosing.
Methods

328 inpatients in academic medical center.
• Inclusions: patients with indication for AC (no specific goal)
• Exclusions: active malignancy or alcoholism
• Blinding: Unblinded



Randomized to PG-dosing or standard care based on
10 mg nomogram.
Primary endpoints were time to first INR in
therapeutic range, time to first INR > 4.0 and time
INR was supratherapeutic.
Secondary endpoints were average warfarin dose and
bleeding events.
Results



Pts with VKORC1 A/A
haplotype had a
significantly faster time
to a therapeutic INR and
a faster time to
supratherapeutic INR.
CYP2C9 genotype had
no effect on time to first
therapeutic INR, but...
Non-wt CYP2C9
genotypes had faster
time to first
supratherapeutic INR.
Genetic Testing: Summary




VKORC and CYP2C9 variants clearly influence
warfarin dose.
The effects of genetic variation are seen during
initiation of therapy, but the longer term effects are not
as clear.
Clinical factors have roughly the same influence as
genetic factors and standardization of the dosing
process is beneficial regardless of the algorithm.
The proper role for genetic testing is not well defined
and further investigation will be needed.
The Program in
Hospital Medicine
Daily Vitamin K
Supplementation Improves
Anticoagulant Stability
Rombouts EK, Rosendaal FR, van der Meer
JM. J Thromb Haemost 2007;5:2043-8
Background


Fluctuations in INR are associated with low
baseline vitamin K intake.
Small studies have shown value in stabilizing
INRs when very low dose vitamin K is added to
the diet.
Methods

200 patients enrolled in a Dutch anticoagulation clinic:
• Inclusions: age 18-80, on AC for at least 1 year
• Exclusions: Renal or hepatic disease, pregnancy, noncompliance.



Randomized to 100 mcg vitamin K daily or placebo.
Primary endpoint was time in therapeutic range.
Secondary endpoints was maximal stability (pts with
INRs in therapeutic range for the entire study period.)
Results


Time in range was
89.5% for the vit K
group and 85.5% for
the placebo group
4% difference (-0.3 8.0).
43% of the vit K
group vs. 24% of the
placebo group had
maximal stability,
RR= 1.8 (1.1-2.7).
Optimal Duration of
Anticoagulation Therapy

Pts who have VTE in the setting of transient risk
factors (immobility, surgery, trauma, etc) have a
low risk of recurrence after appropriate
anticoagulation – 3-6 months.
• 5 yr risk of recurrence after provoked VTE 8-12%

Pts who have VTE in the setting of long-term
risk factors (hereditary thrombophilia, cancer,
lupus anticoagulant, etc) have a high risk of
recurrence after appropriate anticoagulation 6-12
months.
• 5 yr risk of recurrence after unprovoked VTE = 25%
Optimal Duration of
Anticoagulation Therapy, cont’d


Pts who have unprovoked VTE also have an
increased risk of recurrence.
Current guidelines for unprovoked VTE are for
use of VKA for 6-12 months and to consider
indefinite AC.
• Does everyone with unprovoked VTE need
indefinite AC?
• Can we predict who needs indefinite AC?
• How to determine length of therapy?
D-Dimer Testing




608 pts w/unprovoked
VTE and at least 3
months AC.
223 w/abnormal Ddimer level 30 days after
discontinuation of AC.
Half were restarted on
AC, half were not.
Pts w/elevated D-dimer
were more likely to have
recurrence if not on AC.
Palateti G, Cosmi B, Legnani C, et al. N Engl
J Med 2006;355:1780-9.
Ultrasonography

Cohort study of 313
pts in Italy.
• 80 thrombophilia
• 124 unprovoked
• 109 provoked



3 months of AC.
Serial U/S at 3, 6, 12,
24 and 36 months.
Pts w/residual clot on
U/S were at much
higher risk of
recurrent VTE.
Prandoni P, Lensing AWA, Prins MH, et al. Ann Intern
Med. 2002;137:955-560
Recommendations


Provoked VTE: 3-6 months, depending on the
severity of the event.
Unprovoked VTE: 6-12 months AC, followed
by:
• D-Dimer OR venous doppler
• If negative, can discontinue AC
• If positive, continue AC and reassess periodically


Chest, 2004 recommends lifelong AC as a
reasonable alternative for unprovoked VTE.
Recurrent VTE: lifelong AC.
Recommendations, cont’d

VTE and Thrombophilia:
• APLA, ATIII or Protein C + S def’cy: lifelong AC
• Factor V Leiden AND Prothrombin Gene
mutation 20210: lifelong AC or treat as unprovoked
VTE.
• Factor V Leiden homozygote: lifelong AC or treat
as unprovoked VTE.
• Factor V Leiden heterozygote OR Prothrombin
Gene mutation 20210: treat as unprovoked VTE.
Conclusions

Anticoagulation clinics are likely to become more
common as time goes on.
• Standardization alone can bring significant improvements

Genetic testing is not ready for prime time.
• Will likely be used in the near future
• Effects likely seen in initiation of therapy


Low dose vitamin K can help stabilize hard to control
patients.
Use D-dimer or venous ultrasonography to identify
patients at high-risk for recurrent VTE and consider
longer duration of AC therapy.