CE: Alpana; MBC/200642; Total nos of Pages: 4;
MBC 200642
Original article 1
Factor X and factor II activity levels do not always agree in
warfarin-treated lupus anticoagulant patients
Terry K. Rosborough, Jennifer M. Jacobsen and Michele F. Shepherd
Objective Warfarin therapy is used in lupus anticoagulant
patients with thrombosis and yet the prothrombin time
(PT)/international normalized ratio (INR) in these patients
can sometimes be falsely elevated. Both a PT-based factor II
(FII) assay and a chromogenic, enzymatic factor X (CFX)
assay have been used for monitoring when the INR may be
artifactual. This study compared FII and CFX assays in lupus
anticoagulant-positive and lupus anticoagulant-negative
warfarin-treated patients.
Methods Cross-sectional study of samples from 21 lupus
anticoagulant-positive and 19 lupus anticoagulant-negative
outpatients. Plasma samples were simultaneously
measured for FII and CFX and the ratio of FII/CFX was used
to measure concordance.
Main results Compared with lupus anticoagulant-negative
patients 14 of the 21 lupus anticoagulant-positive patients
had lower FII/CFX ratios (P < 0.01). Three of the patients
had ratios less than 0.6 indicating strong disagreement
(P < 0.0001). The patient with the lowest FII/CFX ratio
had evidence suggesting a specific antibody to FII.
Another patient showed that the discordance between
Introduction
Lupus anticoagulant is an antiphospholipid antibody that
may be present in a variety of diseases and may predispose patients to thrombosis that requires warfarin treatment [1]. Warfarin therapy is usually monitored using the
international normalized ratio (INR), which is derived
from the prothrombin time (PT) clotting test. In some
patients and laboratories, the lupus anticoagulant may
falsely elevate the INR [2]. Some laboratories and clinicians use a PT-based factor II (FII) assay to monitor
warfarin therapy when an INR artifact is suspected [3].
However, because a lupus anticoagulant can affect the
PT it may also affect the FII assay. In addition, specific
FII antibodies can occur in lupus anticoagulant patients
that could cause a falsely low measurement of FII [4–6].
In some institutions, an enzymatic, chromogenic factor X
(CFX) assay that avoids clotting test artifacts is used for
alternative warfarin monitoring [2,7].
Both the FII and CFX assay results are reported as a
percentage of normal factor activity, with pooled normal
plasma having 100% activity. The therapeutic range for
these assays, corresponding to a therapeutic INR in
lupus anticoagulant-negative patients is approximately
20–40% [2,8,9]. We compared the results of CFX and FII
in warfarin-treated lupus anticoagulant-positive and
0957-5235 ß 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins
FII and CFX varied over time. The CFX assay in the
laboratory was technically superior, more precise, and
less costly.
Conclusion The CFX assay is preferred for warfarin therapy
monitoring in lupus anticoagulant patients when INR
artifacts are suspected. Blood Coagul Fibrinolysis 21:000–
000 ß 2010 Wolters Kluwer Health | Lippincott Williams &
Wilkins.
Blood Coagulation and Fibrinolysis 2010, 21:000–000
Keywords: chromogenic factor X, factor II, international normalized ratio,
lupus anticoagulant, warfarin
Medical Education Department and Coagulation Laboratory, Abbott
Northwestern Hospital, Minneapolis, Minnesota, USA
Correspondence to Michele F. Shepherd, PharmD, Medical Education Office
11313, Abbott Northwestern Hospital, 800 East 28th Street, Minneapolis,
MN 55407, USA
Tel: +1 612 863 5149; fax: +1 612 863 3672;
e-mail: [email protected]
Received 8 December 2008 Revised 3 November 2009
Accepted 18 November 2009
warfarin-treated lupus anticoagulant-negative patients
to see if the results were interchangeable for monitoring.
Methods
This study was reviewed by the Institutional Review
Board and met the criteria for exempt review.
Patients
All patient care was at the discretion of each patient’s
physician. In one hematology clinic it was the practice to
monitor all lupus anticoagulant patients taking warfarin
with both the CFX and FII assays. We identified 21
patients from this practice who were monitored serially
by our reference laboratory. Patients were monitored for
a median of 10 months (range 2–36 months) and had a
median of 18 paired CFX and FII measurements (range
4–57). No other information about these patients was
available to us. All patients had been diagnosed with
lupus anticoagulant but not all lupus anticoagulant assays
were performed in our laboratory.
For comparison, we selected routine therapeutic INR
blood samples from 19 chronic, stable warfarin patients
from two of our nonhematology clinics and confirmed that
each of these samples was lupus anticoagulant negative;
CFX and FII were measured on each of these samples.
DOI:10.1097/MBC.0b013e32833581a3
CE: Alpana; MBC/200642; Total nos of Pages: 4;
MBC 200642
2010, Vol 00 No 00
Blood samples were collected directly into 0.105 mol/l
(3.2%) sodium citrate coagulation tubes (BD Vacutainer;
Becton Dickinson Company; Franklin Lakes, New
Jersey, USA), which were centrifuged at 120 s at
8500 rpm using the STAT Spin centrifuge (STAT Spin;
Inc.; Norwood, Massachusetts, USA). Plasma was sampled
in the coagulation analyzer directly from the centrifuged
tubes to minimize platelet activation. Although CFX
testing was available at all times, some samples were
processed into platelet-poor plasma and frozen for the
FII assay, as it was only performed during the day.
Confirmatory lupus anticoagulant testing was performed
using the Staclot lupus anticoagulant assay (Diagnostica
Stago, Asnieres-sur-Seine, France) on a Stago Start four
and the lupus anticoagulant SURE dRVVT assay (Precision Biologic, Dartmouth, Nova Scotia, Canada) on a
Stago Compact instrument. If either one of these confirmatory tests was positive, a patient was considered to
be lupus anticoagulant-positive.
AQ1
CFX was measured on the STA-R Evolution (Diagnostica Stago; Parsippany, New Jersey, USA) using the
DiaPharma Factor X Kit that uses the amidolytic substrate S-2765 and Russell’s viper venom as an activator
(DiaPharma; West Chester; Ohio, USA). A standard curve
was created using serial dilutions of Unicalibrator reference plasma (Diagnostica Stago; Parsippany, New Jersey,
USA) containing a known amount of factor X. The
resulting optical density is directly proportional to the
amount of factor X. The CFX assay in our laboratory had
a coefficient of variation of 2–3%. The lowest reportable
CFX result is 5%.
FII assays were performed on the STA-R Evolution using
a modified PT assay with Neoplastin CI plus reagent
(Diagnostica Stago; Parsippany, New Jersey, USA). A
standard curve was created using serial dilutions of Unicalibrator reference plasma (Diagnostica Stago; Parsippany, New Jersey, USA) containing a known amount of
FII. The patient sample was mixed with commercially
available FII-deficient plasma (Precision Biologic;
Dartmouth, Nova Scotia, Canada) and the resulting PT
was compared against the standard curve. The degree of
correction of the PT was proportional to the activity of
FII in the patient’s plasma. The FII assay in our laboratory had a coefficient of variation of 3–6%. The lowest
reportable FII result was 7%.
commercially available FVII-deficient plasma (Precision
Biologic; Dartmouth, Nova Scotia, Canada) and the
resulting PT was compared against the standard curve.
The degree of correction of the PT was proportional to
the activity of FVII in the patient’s plasma. The FVII
assay in our laboratory had a coefficient of variation of
3–6%. The lowest reportable FII result was 7%.
Data analysis and statistical evaluation
Data analysis was with JMP 7 (SAS Institute; Cary, North
Carolina, USA). Comparison of group data was with a
nonparametric method (Wilcoxon rank sum).
Results
We used the ratio of FII to CFX as the primary measure
of assay concordance with a ratio below 1.0 indicating a
sample in which the FII was lower than the CFX.
Figure 1 shows the distribution of FII/CFX ratios for
all patients. The lupus anticoagulant-negative patients
had a median ratio of 0.96, a range of 0.58–1.68, and the
middle 80% of values between 0.73–1.22. Fourteen of
the 21 lupus anticoagulant-positive patients had lower
FII/CFX ratios (P for difference < 0.01). Six patients had
median ratios less than 0.75 (P for difference < 0.001)
with three of the ratios under 0.6.
Figure 2 shows the FII/CFX ratios for patient 03 who had
39 paired measurements over 24 months with a median
FII/CFX ratio of 0.32. The sudden peak in the ratio
occurred when the patient was not on warfarin for a short
time. On the last sample in this patient’s series the lupus
anticoagulant was confirmed as positive and the patient
had a CFX of 29%, a FII of 13%, and a FVII of 30%. As
both the FII and FVII assays are based on the PT, the
discrepancy between these two factors in this patient
infers that the low FII result was not due to general
Fig. 1
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Factor VII assays (FVII) were performed on the Stago
Compact using a modified PT with Neoplastin CI plus
reagent (Diagnostica Stago; Parsippany, New Jersey,
USA). A standard curve was created using serial dilutions
of Unicalibrator reference plasma (Diagnostica Stago;
Parsippany, New Jersey, USA) containing a known
amount of FVII. The patient sample was mixed with
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
LA-
Laboratory assays
II:CFX ratio
2 Blood Coagulation and Fibrinolysis
ID code
Distribution boxplots of FII/CFX ratios in lupus anticoagulant-positive
patients and lupus anticoagulant-negative controls. The line across the
middle of each box identifies the median value, the box identifies the
interquartile range (IQR), and the ‘whiskers’ identify the range extending
an additional 1.5 IQR.
CE: Alpana; MBC/200642; Total nos of Pages: 4;
MBC 200642
Factor X and factor II Rosborough et al. 3
Fig. 2
lupus anticoagulant-positive patients in this study had
lower FII/CFX ratios with six of the 21 patients having
discrepancies in FII that would clearly impact their
warfarin dose.
1.2
II:CFX ratio
1
0.8
0.6
0.4
0.2
0
0
5
10
15
20
25
Months
Variation of FII/CFX ratio over time in patient 03.
interference of the lupus anticoagulant with the PT assay
but to something specific for FII.
Figure 3 shows the FII/CFX ratios for patient 04 who had
47 paired measurements over 11 months. There were two
distinct periods when the FII assay became discrepant.
We do not know anything about the treatment of this
patient over the time period. On the last sample in this
patient’s series the lupus anticoagulant was negative and
all three assays were concordant with a CFX of 29%, a FII
of 35%, and a FVII of 35%.
Discussion
Fig. 3
1.2
1
II:CFX ratio
AQ2
Activity of factors II and X are reduced by warfarin
treatment but because they have different half-lives in
the body, they may not always be at the same level at any
point of time. The results in warfarin-treated, lupus
anticoagulant-negative patients in this study showed
that as the ratio of FII/CFX centered on 1.0, there was
a variability that ranged from 0.75 to 1.25 in about 80% of
the patients. In contrast, the majority of warfarin-treated,
There are two main possibilities for this discrepancy. The
first is that the lupus anticoagulant itself interfered with
the PT assay used to measure FII. The second possibility
is that some lupus anticoagulant patients had factor II
antibodies (4–6). In patient 03 in this study, who had the
greatest discrepancy between FII and CFX, the finding
that the FVII result, which is based on the same PT assay
used to measure FII, agreed with the CFX suggests that
the patient had a FII antibody. In contrast, patient 04
demonstrated concordance of all three assays on a final
plasma sample. However, the concordance between FII
and CFX in this patient varied remarkably over time.
Because we did not measure FVII in all lupus anticoagulant-positive patients we cannot say that all discrepant FII and CFX results were due to FII antibodies.
The lupus anticoagulant itself may have lowered the FII
results in some patients. We also did not have serial lupus
anticoagulant results on patient samples, so we could not
correlate the strength of lupus anticoagulant positivity
with the FII/CFX discrepancy.
There are several technical advantages in our laboratory
favoring the CFX assay over the FII assay. Not only is the
precision of the CFX assay superior to that of the FII
assay but also the stability of the CFX reagents greatly
exceeds that of the FII reagents. This allows us to quickly
offer the CFX assay at all times, something we cannot do
with the FII assay. Also, the overall cost of performing the
CFX assay is less than that of the FII assay.
In conclusion, the PT-based FII assay produces importantly lower results than the enzymatic CFX assay in
many lupus anticoagulant patients. This may be due
either to additional antibodies to FII or to a general
effect of the lupus anticoagulant on the FII assay. In
the laboratory, the CFX assay is superior from a technical
and cost standpoint. The CFX assay is preferred over the
FII assay when evaluating warfarin therapy in lupus
anticoagulant patients who may have INR artifacts.
References
0.8
1
0.6
2
0.4
3
0.2
4
0
0
2
4
6
8
Months
Variation of FII/CFX ratio over time in patient 04.
10
12
5
Baker WF, Bick RL. The clinical spectrum of antiphospholipid syndrome.
Hematol Oncol Clin N Am 2008; 22:33–52.
Moll S, Ortel TL. Monitoring warfarin therapy in patients with lupus
anticoagulants. Ann Intern Med 1997; 127:177–185.
Trask AS, Gosselin RC, Diaz JA, Dager WE. Warfarin initiation and
monitoring with clotting factors II, VII, and X. Ann Pharmacother 2004;
38:251–256.
Bardin N, Alessi MC, Dignat-George F, et al. Does the antiprothrombin
antibodies measurement provide additional information in patients with
thrombosis? Immunobiology 2007; 212:557–565.
Tincani A, Morozzi G, Afeltra A, et al. Antiprothrombin antibodies: a
comparative analysis of homemade and commercial methods. A
collaborative study by the Forum Interdisciplinare per la Ricerca nelle
Malattie Autoimmuni (FIRMA). Clin Exp Rheumatol 2007; 25:268–
274.
CE: Alpana; MBC/200642; Total nos of Pages: 4;
MBC 200642
4 Blood Coagulation and Fibrinolysis
6
7
8
9
2010, Vol 00 No 00
Galli M, Borrelli G, Jacobsen EM, et al. Clinical significance of different
antiphospholipid antibodies in the WAPS (warfarin in the antiphospholipid
syndrome) study. Blood 2007; 110:1178–1183.
Rosborough TK, Shepherd MF. Unreliability of international normalized ratio
for monitoring warfarin therapy in patients with lupus anticoagulant.
Pharmacotherapy 2004; 24:838–842.
Lind SE, Callas PW, Golden EA, Joyner KA Jr, Ortel TL. Plasma levels of
factors II, VII and X and their relationship to the international normalized ratio
during chronic warfarin therapy. Blood Coagul Fibrinolysis 1997; 8:48–53.
McGlasson DL, Romick BG, Rubal BJ. Comparison of a chromogenic factor
X assay with international normalized ratio for monitoring oral anticoagulation
therapy. Blood Coagul Fibrinolysis 2008; 19:513–517.