|
|
| |
| Ask the Expert |
|
Heparin and Anti-Xa Assays
Bleeding Time
Mixing Studies
Thrombophilia Testing
Factor VIII
Technical Issues
Heparin
and Anti-Xa Assays
Question:
My assignment is to institute new testing in our laboratory
for low molecular weight heparin. What criteria should
one use to choose reagents? What criteria are used to
establish reference curves, considering the different
types of “heparin” (unfractionated heparin,
LMWH, “heparinoids” [direct FXa inhibitors])
available?
Submitted from Florida
Response:
There are numerous manufacturers of test kits for measuring
the anti-Xa activity of the various “heparin”
products that you noted. You may find a list of diagnostic
companies under “Links” on our web site. The
majority of kits available in the US measure anti-Xa
activity of heparins using a chromogenic substrate.
For LMWH one can also use kits that measure anti-IIa
activity. Some manufacturers have kits specifically
for unfractionated heparin (UFH) and LMWH whereas others
can measure both using one kit. The calibration issue
is critical. For LMWH, manufacturers generally provide
a standard in their kits that most often is Fragmin
(the “gold standard”). For UFH it is best
to use the heparin that your pharmacy is administering.
Naturally there are a number of practical drawbacks
with this. For “heparinoids”, better known
as direct FXa inhibitors it would be best to use the
product in question to establish the standard curve.
However to date there is limited practical knowledge
in this area. Controls for Anti-Xa kits are generally
provided by the manufacturers, however for therapeutic
agents other than UFH and LMWH you may need to prepare
your own controls (one should be subtherapeutic and
the other within the therapeutic range).
The greatest difficulty with these assays is fulfilling
requirements for test validation. You will need to show
that your test system can discriminate anti-Xa (or anti-IIa)
activity at various levels (no heparin on board, borderline
[subtherapeutic], or therapeutic). Likewise you will
need to demonstrate that your assay is accurate. The
US Pharmacopeia has UFH standards at various concentrations
that you may consider using for this purpose.
Question:
Our Quality Committee met today, and one subject
of conversation was regarding the use of an Anti-Xa
assay for monitoring unfractionated heparin (UFH). Our
Pathologist suggested that Anti-Xa monitoring was the
new standard of care, replacing the APTT altogether.
I would comment that the Clinical Guidelines (AHA, Chest,
etc) have mentioned the test but have not provided definitive
guidelines on application to practice. The questions
we have are as follows:
|
1. |
Is the Anti-Xa test an accepted practice at the
community hospital level?
|
|
2. |
Is this viewed as an acute test with rapid turn
around or is a small delay associated with send-out
appropriate?
|
|
3. |
Pharmacy perspective: We treat many elderly patients
with varying degrees of renal dysfunction. We recognize
that Lovenox dosing requires adjustment for CrCl
< 30 ml/min, but we have no guidelines from the
company. We have also noticed physicians doing off-label
dose adjustment (Lovenox 1mg/kg Q24H (Label is Q12H).)
There is no literature support of this, and an Anti-Xa
activity would provide guidance to avoid therapeutic
failures with this LMWH. Furthermore we anticipate
that Anti-Xa activity will eventually be a standard
of care and improve patient safety/decrease bleeding
risk for both LMWH and UFH. At present, we convert
all Anti-Xa orders to UFH with APTT follow-up. In
this case, I would anticipate that send out of the
lab is perfectly appropriate.
|
Submitted from Florida
Response:
In answer to your first question, the test is generally
not performed at the community hospital level. The caveat
for all your questions is the following: at this time
point the APTT is the accepted standard of practice for
monitoring unfractionated heparin. It would be too costly
to monitor UFH using an Anti-Xa assay. Therefore the APTT
is the test of choice. Naturally one cannot use the APTT
for monitoring LMWH so the Anti-Xa is one's choice (or
an Anti-IIa assay).
At the root of all your questions is a CAP "recommendation"
that one must be able to show how an institution's APTT
compares with the Anti-Xa assay. In other words if your
reagent's therapeutic UFH range for APTT is 45-75 seconds
(example only) would this fall into the therapeutic window
of 0.3-0.7 Anti-Xa Units/ml? One can ascertain this by
testing at least 50 samples from actual patients receiving
known doses of therapeutic heparin. Both APTT and Anti-Xa
testing would be performed on each of these samples. Following
data collection, plot Anti-Xa (x-axis) versus APTT (y-axis)
and by linear regression compute a line of best fit. From
this one can determine what APTT corresponds to 0.3 U/ml
and what upper APTT corresponds to 0.7 U/ml. The caveat
is that the standard curve used for the Anti-Xa assays
should be prepared with the heparin that will be used
in your institution (sequester a lot of heparin in the
pharmacy ~12 months). Likewise this holds only for a particular
lot number of APTT reagent (sensitivities change from
lot to lot) and would need to be reassessed with each
new lot of APTT reagent and/or heparin.
Concerning turn-around-time for the Anti-Xa assay, it
should be similar to an APTT. Afterall the assay, like
the APTT is used for the purpose of monitoring a drug
and potentially to make dosage adjustment. As to your
final points raised by your pharmacy, I have found that
physicians prefer to "monitor" to some degree
any LMWH given to the following patients: obese, pregnant,
pediatric, or with renal dysfunction. Once again since
appropriate dosing is critical I would not "send
out" the test.
back
to the top
|
| Bleeding
Time
Question:
Should Bleeding Times be performed on infants? Is there
a reference range? How often does one wick? I would
very much appreciate any comments.
Submitted from Pennsylvania
Response:
The Bleeding Time test lacks
sensitivity and specificity. Furthermore many factors
such as low platelet count, low hematocrit, and aspirin
or aspirin containing products can confound results.
Add to that numerous analytical variables such as location
of the incision, depth of incision, skin thickness,
orientation of the blade, and so forth. All of these
considerations added to the inherent problems of dealing
with a newborn. However, taking all of these points
into consideration some physicians still find clinical
utility in performing this assay in the newborn. A caveat
is test interpretation since it is difficult for the
average institution to establish a newborn or infant
reference range. You may wish to go to PubMed and initiate
a search (bleeding time test in newborns) to read the
experiences of others on both side of the issue. You
may also wish to do a search on www.google.com using
bleeding times and newborns or bleeding time devices
as a way to search for more information. Finally, visit
the web sites of several companies that manufacture
bleeding time devices for infants: www.itcmed.com (ITC:
Surgicutt), www.helena.com (Helena Laboratories: Triplett
Device, Tip Tripper), and www1.fishersci.com (Fisher:
Simplate-Organon).
Your questions regarding methodology would be answered
in respective bleeding time device product inserts.
Also please refer to NCCLS document Approved Guideline
H45-A, 1998: Performance of the Bleeding Time Test wherein
your question regarding wicking is addressed (every
30 seconds).
back
to the top
|
| Mixing
Studies
Question:
How are mixing studies performed?
Submitted from California
Response:
A simple question with innumerable answers! To date
there does not seem to be a consensus as to how to perform
and interpret a mixing study. From my own practical
experience here are some suggestions.
|
1. |
It is critical that you use plasma that has been
pooled from at least 30 normal donors. For PT/APTT
mixing studies I would suggest using an in-house
prepared pooled normal plasma (NPP): aliquot and
freeze at -70 0C (vials thawed in a 37
0C heating block for 10 minutes-stability:
8 hours at 2-8 0C). If this is not possible
then one can purchase pooled normal plasma from
several vendors. However, you must be absolutely
sure that the PT and APTT of those purchased plasmas
fall within your PT and APTT reference ranges.
|
|
2. |
If PT is prolonged by more than +2SD, perform
a mixing study. Mixing Study with 50% normal pooled
plasma: (200 ul plasma + 200 ul NPP). Repeat PT
assay on this mix. Failure to correct the PT to
within the reference range (not everyone’s
criteria!) with 50% NPP indicates the presence of
an inhibitor. Correction by NPP should be seen with
factor deficiencies or effects due to warfarin.
|
|
3. |
If APTT is prolonged by more than +2SD, I would
suggest to perform the following:
| a) |
Perform a Thrombin Time pre and post treatment
with a heparin neutralizing agent to rule
out the presence of heparin.
|
| b) |
Mixing study with 50% NPP: (200 ul plasma
+ 200 ul NPP). Repeat APTT assay on this mix.
Lack of correction with mixing study (failure
to correct APTT to within reference limits-again
not everyone’s criteria) with 50% NPP
indicates presence of an inhibitor. Correction
by normal plasma should be seen with factor
deficiencies or effects due to therapeutic
heparin.
|
|
Testing can be performed immediately upon mixing
(PT and/or APTT) and following incubation (APTT only).
Incubation is used to tease out time and temperature dependent
inhibitors of which the most common is a Factor VIII inhibitor.
For the APTT Mixing Study one can use the following mixtures:
100% NPP (only NPP), 50% NPP (equal parts NPP and patient
plasma), 25% NPP (1 part NPP and 3 parts patient plasma),
and 0% NPP (patient plasma only). These mixtures can be
tested immediately and after incubation at 370C
for 2 hours. At the end of 2 hours one should also not
only measure the 100% NPP and 100% patient plasma but
a 50/50 mix of these two (incubated separately) that serves
as a “control” for FV and FVIII lability. A
greater than 15 second difference between the 50/50 mix
(50% NPP tube) that incubated together for 2 hours and
the 50/50 "control" suggest the presence of
a time/temperature dependent inhibitor.
As noted, interpretation of mixing studies can be more
complex than actually performing the tests. While it can
be appreciated that laboratories need to draw a line in
the sand so to speak (correction versus no correction),
one must make that assessment within the context of the
baseline APTT. Certainly an APTT of >80 seconds in
a FIX deficient patient that shortens to within 2 seconds
of an upper limit does not herald an inhibitor.
Question:
What are your thoughts on using a fresh normal coagulation
control in place of normal pooled plasma (NPP) for mixing
studies? The information you provide will be beneficial
in updating our procedures. Concerning APTT mixing studies
with incubation, we incubate for 1 hour. There has been
some discussion about whether to continue past the one
hour incubation time. I don't recall noticing differences
between the 1 hr and 2 hr incubation results. What is
you opinion on this?
Submitted from Kansas
Response:
In response to your question about the use of a normal,
lyophilized control versus NPP to be used for mixing studies,
I have a question for you: how many patient samples do
you receive that are lyophilized? The point here is that
one should perform mixing studies with a "like"
product. Also you must appreciate that lyophilized control
plasmas have been "tweaked" for their purposes
and may contain certain stabilizers.
As to your second question regarding incubation times.
I would argue that since a 2 hour incubation time is used
to demonstrate the inhibitory effect of a FVIII inhibitor
in an Bethesda assay that this time period should also
be use for the screening APTT mixing study. If incubation
time is less than 2 hours one may miss time/temperature
dependent FVIII inhibitors because the kinetics of most
FVIII inhibitors require at least 2 hours in order to
demonstrate their effect. One could incubate longer but
that adds more time to the assay than is really necessary.
back to the
top
|
| Thrombophilia
Testing
Question:
I just received a request for a custom hypercoagulability
panel from a physician in one of our hospitals. The
following tests were included: Antithrombin III, Anticardiolipin
antibodies, Protein C activity and antigen, Protein
S (activity, Total & Free), Lupus Anticoagulant,
Factor V Leiden, and beta2 glycoprotein I (IgG, IgM).
From what I have read, there seems to be little agreement
about what the best screening panel should be. Any thoughts?
Submitted from Florida
Response:
You are correct, there is little
consensus as to what to order. It is subject to interpretation
even amongst the experts (for an excellent perspective
refer to Bauer, Rosendaal, Heit. Hypercoagulability:
Too Many tests, Too Much Conflicting Data. Hematology
2002, American Society of Hematology Education Program
Book). For a hypercoagulability workup it is important
that a good history (personal and familial) be obtained.
Also is the thrombotic event venous or arterial in nature?
Certainly the battery you listed above with the exception
of the ACA, LA, and beta 2 GPI (also implicated in arterial
thrombosis) would be more appropriate for venous thrombosis.
A strong family history would prompt ordering of assays
for inherited thrombophilias (FV Leiden, PT20210, Antithrombin,
Protein C, Protein S) whereas Lupus Anticoagulant, ACA,
beta2-GPI are generally considered as acquired hemostatic
causes for thrombosis. Unfortunately many physicians
do not have enough time to get a good history and have
only a “one shot” chance to see the patient
so they order everything. Even in light of the latter,
I would not order a PC antigen, PS Total or Free antigen
unless functional testing was abnormal. Should these
tests be needed, they can be add on orders and not require
a patient to be seen again. Likewise there is considerable
controversy regarding ACA versus beta2-GPI but the Lupus
Anticoagulant Scientific and Standardization Committee
(SSC) of the ISTH still has not made recommendations
as to which/both to use. The final question: when all
these tests are normal, how esoteric does one go from
here?
back
to the top
|
| Factor
VIII
Question:
Do you know where I can get a method and some information
on the two-stage factor VIII assay? Does anyone in the
US do this test anymore?
Submitted from Wisconsin
Response:
I do not know if anyone in the
US is doing a two-stage FVIII assay. You may wish to
go to PubMed and do some searching since the two-stage
seems to be coming into the light of day again particularly
in Europe. There are some hemophilia A patients with
particular mutations who will give normal results with
the one-stage assay but quite abnormal with the two
stage. There are numerous references on this topic when
searching on the PubMed site at www.ncbi.nlm.nih.gov
back
to the top
|
| Technical
Issues
Question:
How can one convert a plasma sample to serum?
Submitted from New York
Response:
You can add some thrombin to the plasma and wait for
clot formation. Usually 1 U/ml thrombin works but a
higher concentration should not be problematic. Besides
the visual appearance of a clot, you could perform a
Fibrinogen or Thrombin Time test to be absolutely certain
that all fibrinogen has been removed.
back
to the top
|
|
| |
| |
| |
| |
| |
| |
|
