Received: 20 August 2019 Revised: 24 October 2019 Accepted: 28 October 2019

DOI: 10.1002/ajh.25669

TEST OF THE MONTH

Coagulation mixing studies: Utility, algorithmic strategies and

limitations for lupus anticoagulant testing or follow up of
abnormal coagulation tests

Emmanuel J. Favaloro
Department of Haematology, Institute of
Clinical Pathology and Medical Research
(ICPMR), Sydney Centres for Thrombosis and
Haemostasis, NSW Health Pathology,
Westmead Hospital, Westmead, New South
Wales, Australia
Correspondence
Emmanuel J. Favaloro, Department of
Haematology, Institute of Clinical Pathology
and Medical Research (ICPMR), Westmead,
New South Wales, 2145, Australia.
Email: emmanuel.favaloro@health.nsw.gov.au
Abstract
Coagulation testing underpins the investigation of hemostasis and/or monitoring of
anticoagulation therapy for prevention and/or treatment of thrombosis related
pathology. Assessment of coagulation results requires comparison against a normal
reference range or interval (NRR/NRI). Results flagged as “abnormal” (ie, above the
NRR/NRI for patients not on anticoagulant therapy), typically require further evalua-
tion. eg, follow up or reflexive testing is used to identify the reason for prolongation,
especially when supported by clinical context (eg, bleeding). Mixing tests may have
utility to help identify the pathway of follow-up testing (ie, towards investigation of
factor deficiencies, or else inhibitors), and are also useful for investigation of lupus
anticoagulants (LA). In general, mixing tests that “correct” tend to suggest the pres-
ence of factor deficiencies, where as those that do not correct suggest the presence
of “inhibitors”. Various approaches can be used to identify correction/non-correction,
and all have strengths and limitations. Furthermore, eventual identification of causal
factor deficiencies or even “inhibitors” may (eg, factor VIII or IX deficiencies or inhibi-
tors) or may not (eg, factor XII deficiency) be clinically important. Ultimately, mixing
studies performed in view of appropriate clinical scenarios (eg, bleeding patient) and
for LA investigations in symptomatic patients will have best utility.

1 | I N T RO D UC T I O N INR; unfractionated heparin therapy for APTT), (c) to evaluate potential

Coagulation testing underpins the investigation of hemostasis and/or
monitoring of anticoagulation therapy for prevention and/or treatment
of thrombosis related pathology. 1,2
The most commonly performed
coagulation tests in our large pathology network, and likely most other
coagulation laboratories, in order of test requests, are (a) prothrombin
time/international normalized ratio (PT/INR), (b) activated partial
thromboplastin time (APTT), (c) fibrinogen, (d) D-dimer, (e) thrombin
time (TT). Basic coagulation tests may be used for a variety of reasons,
but primarily to (a) screen for defects of hemostasis (eg, factor deficien-
cies, either prior to surgery or as a follow up investigation in a patient
being investigated for a bleeding diathesis; to assess possibility of dis-
seminated intravascular coagulation [DIC]), (b) to monitor anticoagulant
therapy (eg, vitamin K antagonist [VKA] therapy such as warfarin for
anticoagulant status in an unknown case (eg, unconscious emergency
admission requiring surgical intervention).1-4 As a summary of this back-
ground, please refer to Table 1, which lists the main tests covered in
this review, as well as test limitations.
Assessment of the results of routine coagulation tests requires
comparison against a normal reference range or interval (NRR/NRI).5
Results within the NRI are said to be “normal”, where as those above
the NRI are said to be “prolonged”; on some, but not all occasions
such results can also be said to be “abnormal”. Thus, for patients being
investigated for hemostasis, either as a “preoperative screen” or for
investigation of some hemostasis defect (eg, factor deficiency), results
above the NRI would be flagged as abnormal and typically require fur-
ther evaluation, for example by clinical follow-up or immediate reflex-
ive testing. Alternatively, for patients being monitored for
anticoagulation therapy, results would be expected to be above the

Am J Hematol. 2020;95:117–128. wileyonlinelibrary.com/journal/ajh © 2019 Wiley Periodicals, Inc. 117

118 FAVALORO

TABLE 1 Overview of main routine coagulation tests and their potential uses

Test name Abbreviation Sensitive to Main uses Limitations

Prothrombin time PT VKAs (eg, warfarin); deficiency in or
inhibitors of factors X, VII, V, II, I
(fibrinogen); DOACs (especially
dabigatran and rivaroxaban, and
to a lesser extent apixaban)
International INR INR = (patient PT/MNPT)ISI
normalized ratio
Activated partial APTT Unfractionated heparin; deficiency
thromboplastin in or inhibitors of factors XII, XI, X,
time IX, VIII, V, II, I (fibrinogen); VKAs;
DOACs (especially dabigatran, and
to a lesser extent rivaroxaban and
apixaban); Lupus anticoagulant
(LA)
Thrombin time TT Unfractionated heparin; direct
thrombin inhibitors (eg,
dabigatran); deficiency in or
inhibitors of factor I (fibrinogen);
FDPs/D-dimer
Russell Viper RVVT Lupus anticoagulants (LA); DOACs
venom time (especially rivaroxaban, to a lesser
extent dabigatran and apixaban)
Assessment of patient hemostasis
(eg, factor deficiency); assessment
of DIC; assessment of liver
disease status; monitoring of VKA
therapy (as INR)
Monitoring of VKA therapy
Assessment of patient hemostasis
(eg, factor deficiency); assessment
of DIC; assessment of liver
disease status; monitoring of
unfractionated heparin therapy;
part of LA testing
Assessment of patient hemostasis
(eg, fibrinogen deficiency);
assessment of DIC; screen for
presence of unfractionated
heparin; screen for presence of
dabigatran
Part of Lupus anticoagulant (LA)
testing
Assay reagent variation for various
sensitivities; normal PTs do not
exclude factor deficiencies;
prolonged PTs do not
necessarily reflect bleeding risk
Accurate INRs require accurate
MNPT and ISI values
Assay reagent variation for various
sensitivities; normal APTTs do
not exclude factor deficiencies;
prolonged APTTs do not
necessarily reflect bleeding risk
Assay reagent variation for various
sensitivities (eg, thrombin
concentration used in assay
may render increased or
decreased sensitivity to
D-dimer)
Assay reagent variation for various
sensitivities (eg, phospholipid
composition and concentration
affect LA sensitivity)

Note: Individual test sensitivities for anticoagulants including DOACs is impacted by the reagent and instrument combination in use and should be
confirmed by each laboratory.
Abbreviations: DOACs, direct oral anticoagulants; FDPs, fibrin(/ogen) degradation products; LA, lupus anticoagulant; MNPT, mean normal prothrombin
time; ISI, international sensitivity index; VKAs, vitamin K antagonists (eg, warfarin).

NRI, and results would be instead be assessed against, or targeted to,
a therapeutic range (TR). This is typically somewhere around 1.5 to
3 times the NRI (eg, INR of 2.0 to 3.0; APTT of around 1.5-2.5×
normal, or more specifically, the APTT range corresponding to an anti-
FXa level of 0.3-0.7 U/mL). 3,4,6,7
Thus, results for such anticoagulated
patients would only be flagged or regarded as “abnormal” if they
algorithmic follow up (eg, an abnormal PT and APTT may cause reflex
testing of fibrinogen and/or TT and/or D-dimer).9 Note that individual
test sensitivities for anticoagulants including DOACs is impacted by
the reagent and instrument combination in use, and should be con-
firmed by each laboratory. An example of potential reflex testing
(incorporating mixing studies) is shown in Figure 1.

exceeded the TR.

Follow up of abnormal tests depends on the specific situation. For
patients on anticoagulant therapy, the typical response to a result
above the TR is withholding an anticoagulant dose for a period of
time, or reducing the anticoagulant dose, with the aim to obtain a
result within the TR on the next test occasion. For patients not on
anticoagulant therapy, or not known to be on anticoagulant therapy
(eg, unconscious emergency admission), an abnormal coagulation test
result may or may not reflect an important hemostatic defect or defi-
ciency. Nevertheless, since it is feasible that such patients will have a
clinically significant defect (until proven otherwise), it is usually incum-
bent on the laboratory/patient carer/clinician to investigate and
potentially explain the abnormality. Different laboratories and clini-
cians may have different strategies for such patient test follow up, but
algorithmic reflex testing, potentially including mixing studies, pro-
vides one standard and logical approach to facilitate discovery, if per-
mitted by the local regulatory environment and fiscal contraints.8,9
Some of the routine tests themselves (Table 1) may form part of an
2 | MIXING TESTS - WHAT THEY ARE, HOW
T H E Y H E L P P A T I E N T I N V E S T I G A T I O N S , A ND
T H E I R L I M I TA T I O N S
Mixing tests, as the term infers, represents tests performed on a mix-
ture of patient plasma and normal plasma.10,11 Usually, this normal
plasma is a normal plasma pool (NPP). An ideal NPP would comprise:
(a) a minimum of 20 normal individuals (preferably equal ratio male/
female) in order to statistically yield >80 U/dL of each coagulation
factor; (b) platelet count for all individual plasmas and the resultant
pool should be <10 × 109/L; (c) absence of LA in the NPP should be
documented; (d) NPP can be frozen or lyophilized, but commercial
sources should be stored according to manufacturer instructions.
Mixing tests are indicated when the patient's routine coagulation
test results are unexpectedly prolonged (or “abnormal”). Mixing tests
are usually performed as an “immediate mix”, meaning test plasma is
mixed with normal plasma and immediately retested. Special
FAVALORO 119

F I G U R E 1 Mixing studies as part of an algorithmic approach to investigation of unexpected abnormal routine coagulation test clotting times
(CT). Three broad scenarios are feasible: prolonged PT only, prolonged APTT only, or prolonged PT and APTT. Mixing studies are performed using
the test that gave the prolonged CT, and mixing patient plasma 1:1 with normal plasma. Additional testing may be recommended depending on
clinical context and at clinical discretion. The thrombin time (TT) is also helpful for cases of abnormal APTT (the TT should be prolonged if the
prolonged APTT is due to heparin, dabigatran or fibrinogen deficiency; also, an abnormal TT-mix suggests the heparin or dabigatran and a
corrected TT-mix suggests fibrinogen deficiency). Different possibilities exist to explain the prolonged CT in each scenario, and the mixing test CT
result helps direct what further testing may be useful. In the case where both PT and APTT are abnormal, mixing tests will typically align (ie, both
correct or both do not correct); however, it is possible for one test to correct and the other not to correct, dependent on why the original CTs
were abnormal, and/or the method used to identify correction. For example, heparin just above the therapeutic range may lead to grossly
abnormal APTT and mildly prolonged PT if the PT reagent contains a heparin neutralizer. Mixing may then generate a non-correction in APTT, but
a correction in PT if the heparin level is diluted to within the PT reagent's heparin neutralization range. APTT, activated partial thromboplastin
time; PT, prothrombin time; LA, Lupus anticoagulant; DIC, disseminated intravascular coagulation; DOACs, direct oral anticoagulants; VK, vitamin
K; VKA, vitamin K antagonist [Color figure can be viewed at wileyonlinelibrary.com]

consideration is, however, required for some situations, especially in
the case of FVIII inhibitors, which may only become apparent upon
“incubated mixing” (see Section 6).
Mixing tests have no utility where patient tests are normal, or in
general when patients are known to be under anticoagulation therapy.
In the former, no additional information can be gained by mixing nor-
mal plasma with a normal patient sample - the result will remain nor-
mal. In the case of known anticoagulation therapy, no additional
information will normally be gained by mixing normal plasma with a
prolonged patient test sample due to the therapy. The outcome is
predicable based on the anticoagulation being applied for therapy
(correction with VKA therapy; non-correction with direct oral antico-
agulants [DOACs] and other anticoagulants such as heparin; etc). An
exception, however, may be made when a patient is suspected of
being on an anticoagulant, but the specific therapy is unknown. In
such cases, mixing studies may form part of the strategy to identify
what subsequent specific testing should be preferentially initiated (eg,
Figure 1). Another exception is in the specific investigation of lupus
anticoagulant (see Section 4).
Mixing tests may utilize different mixtures of patient + normal
plasma, but generally a mixture of 1:1 (or 50:50; i.e., equal volumes) is
applied.9-13 The mixture of patient + normal plasma is subjected to the
same test(s) that showed the initially abnormal patient test result
(Figure 1). The rationale here is that if factor deficiencies are present,
the mixture should generate a normal test result. This is since normal
test results are expected in the normal population, where factor levels
may range from around 50 to 200 U/dL for each individual factor, and
thus reference ranges would reflect this “normal range” of factor levels.
Thus, assuming the normal plasma contains an average of around
100 U/dL of all coagulation factors, then mixing this normal plasma
with a completely factor deficient plasma (even serum) will still provide
a plasma mixture with a composite minimum of ~50 U/dL of each fac-
tor (thereby theoretically sufficient to generate a normal test result).
Evidence for this is shown in Figure 2. Nevertheless, mixing tests work
120
best when single factor deficiencies are present, rather than multiple
factor deficiencies. This is because having multiple factors at levels
close to 50 U/dL in the mix may prolong the clotting time, depending
on the reagent and its factor sensitivity. Thus, when multiple factor
deficiencies are present, the chance of false non-correction increases.
FIGURE 2 Legend on next page.
FAVALORO
The most common examples are PT mixing tests, which might not fully
correct with liver dysfunction, DIC, or vitamin K deficiency, which are
common causes of acquired multiple low factor levels. Such situations
may have additional confounders influencing the ability of the mixture
to clot. eg, serum contains fibrin degradation products (FDPs) which
FAVALORO 121

may interfere with clotting. Vitamin K deficiency (or patients on VKA 3.1 | Mixed test result within the NRI method
therapy) may yield production of “PIVKA” (proteins induced by vitamin
One commonly used method is identifying whether the test/normal
K absence), which may also interfere with clotting. Nevertheless, in the
mixture provides a test result that is now in the NRI, in which case
absence of “inhibitors”, mixtures with such abnormal plasmas will still
“correction” is said to occur. This is the option recommended by the
yield clotting times closer to that of the normal than the original patient
CLSI guidelines for investigation of LA,17 and is also easy to adopt for
result (refer to Figure 2 for some examples).
mixing studies in general. Indeed, this is the option that our laboratory
In contrast, the presence of some “factor inhibitor” (including most
decided to adopt when we developed our complex set of rules for
therapeutic anticoagulants or LA) should not permit generation of a
auto-validation of routine coagulation tests.9,26 The main advantage is
normal coagulation test result in the plasma mixture, because the inhibi-
that the method is easy to apply and verify, and does not require gen-
tor can inhibit the normal plasma as well as the patient plasma. As a
eration of alternate methods or cut-off values requiring additional ver-
general rule of thumb, in our experience, a patient's plasma containing
ification. The main disadvantage, common to all methods, is that it
an inhibitor of coagulation should generate a mixed plasma test result
does not permit 100% differentiation of “factor deficiencies” vs “fac-
close to the average of the normal and patient test result. Naturally, not
tor inhibitors”. The main failure with this method is where the “abnor-
all test mixtures provide ideal test results completely representative of
mal” patient PT or APTT test result is just above the NRI. The likely
expectations, due to assay imprecision and since not all inhibitors
outcome here is that any mixture will generate a result within the
reflect type 1 kinetics.10,11,16 Nevertheless, the above background can
NRI, irrespective of whether the patient plasma contains factor defi-
in general be translated to: (a) a mixture of NPP with abnormal test
ciencies or inhibitors, and due to “genuine” correction in the former
plasma (TP) that generates a normal mix test result usually suggests fac-
and “false” correction in the latter due simply to inhibitor dilution. This
tor deficiencies in the patient plasma, whereas (b) a mixture that still
is why laboratories that use this method often do not perform mixing
generates an abnormal mix test result usually suggests factor inhibitors
tests unless the PT or APTT are more than slightly prolonged (eg,
(potentially including anticoagulants). Some of the caveats to clear dis-
2 seconds for PT, 5 seconds for APTT).
crimination include: (a) source and quality of normal plasma used for
mixing studies - should reflect a validated (pool) plasma that provides
levels for individual clotting factors around 100 U/dL; (b) some inhibi- 3.2 | Index of circulation anticoagulant (ICA)
tors (notably FVIII) are time and temperature dependent and only show
Another commonly employed method is the index of circulation anti-
inhibition using incubated mixing studies16 (see Section 6); (c) the
coagulant (ICA), also known as the “Rosner Index”, named after the
method used to interpret correction (see Section 3); (d) mixing intro-
method's advocate, and initially developed for LA investigation.10,11,18
duces a dilutional effect; thus a 1:1 mix has the potential to mask a mild
The ICA is identified by the formula:
inhibitor. As a summary of this section, please refer to Table 2.

ICA = ½ð1 : 1 mix CT − NPP CT Þ=patient CT  x 100

3 | INTERPRETATION OF MIXING STUDIES
There are actually several methods available to interpret mixing stud-
ies as having either “corrected” (suggesting factor deficiency) vs “not
corrected” (suggesting factor inhibitor).10,11,14-25
where CT = clotting time of the test under investigation, and
NPP = normal pool plasma. The calculation is not complex and can be
built into laboratory information systems (LIS) or middleware or
instrumentation, and then an interpretation applied against a given

F I G U R E 2 Sample scenarios showing the results of mixing studies for PT (prothrombin time) and APTT (activated partial thromboplastin time).
Data previously not published, but extracted from a previously published study.14,15 Various sample types tested by laboratories participating in
an external quality assessment (EQA) program. Laboratories were asked to identify the likely defect represented by the sample, with each
presenting a prolonged PT and/or APTT. PT and APTT results reported by laboratories as > a given value have been adjusted to that value +1 for
the purpose of graphical representation. Laboratories were also free to undertake any additional testing, including mixing studies using their usual
process. Laboratories also interpreted the results of any mixing tests performed. (A) A factor V inhibitor (~5 Bethesda Units/mL) yielding
prolonged clotting times (CT) (y-axes) with both PT and APTT. Mix test results did not show correction to normal values, and interpretations from
laboratories were an approximate equal distribution of “no correction” or “partial correction” (see Figure E). (B) Normal serum created by clotting/
defibrinating normal pool plasma. PT and APTT values invariably yielded maximally prolonged clotting times (CT) (y-axes) with both PT and APTT,
with the values shown reflective of variable “maximal” times as reported and ranging from >60 to >300 seconds. Mix test results showed
correction to normal values in all but three occasions; in these cases, there was suspicion of transcription error rather than failures in mix tests.
Interpretations from laboratories were almost unanimously “correction” (see Figure E). (C) A lupus anticoagulant (LA) positive sample. PT values
were generally normal, but APTT values generally prolonged. Mix test results for APTT generally showed correction towards normal values, but
usually remained “abnormal”. Here, the initial APTT and the mix APTT test result would depend on the sensitivity of the reagent for
LA. Interpretations were mostly reflective of “non-correction” or “partial correction” (see Figure E). (D) A sample prepared by aluminium
absorption and generating a sample that would mimic a vitamin K deficiency or reflect vitamin K antagonist therapy. The PT and APTT values
were invariably prolonged, and mix test results generally showed correction into normal reference ranges, thereby yielding interpretations
reflective of “correction” (see Figure E)
122 FAVALORO

TABLE 2 Mixing tests - a summary of general principles, and benefits vs limitations

General principles
Benefits of mixing tests
Limitations of mixing tests
Perform clotting time (CT) on mixture of patient plasma (when it provides an abnormal CT) plus normal plasma (typically
a pool; “NPP”).
Different mixtures of patient + normal plasma may be used, but most typical is a mixture of 1:1 (or 50:50; i.e., equal
volumes)
Can be applied to most routine coagulation tests, including PT, APTT, TT, as well as to RVVT (eg, as part of lupus
anticoagulant (LA) investigation)
A corrected CT with the mixed plasma typically indicates a factor deficiency/ies in the original patient plasma.
A “non-corrected” CT the mixed plasma indicates a factor “inhibitor” in the original patient plasma. Factor “inhibitors”
can include allo- or auto-antibodies, and anticoagulants such as heparin and DOACs.
Different methods can be used to identify correction vs non-correction, including correction within the normal
reference interval (NRI), index of circulation anticoagulant (ICA), and percent correction method
Provides standard and simple methodology to dichotomize subsequent investigation of abnormal patient test results
for routine coagulation tests
Useful as part of diagnostic strategy for LA investigation
All methods, including correction within the normal reference interval (NRI), index of circulation anticoagulant (ICA),
and percent correction method, will provide effective discrimination and an effective strategy for immediate
follow-up in the vast majority (>95%) of test cases where the method has been validated by the laboratory
Not generally useful as part of follow up of a known anticoagulant therapy effect, except to differentiate VKA therapy
(mixing study correction) vs alternates (mixing study does not normally correct in case of DOACs, or with heparin,
unless mixing brings the heparin level to within the neutralizing ability of a particular reagent)
All methods, including correction within the normal reference interval (NRI), index of circulation anticoagulant (ICA),
and percent correction method, do not provide 100% discrimination/effective strategy for abnormal test follow-up
Additional test costs and operator time, sometimes not reimbursed by funding bodies (i.e., reflex testing may not be
recognized as a specific laboratory test procedure for reimbursement, if not established at the laboratory)

Abbreviations: APTT, activated partial thromboplastin time; DOAC, direct oral anticoagulant; NRI, normal reference interval; PT, Prothrombin time; LA,
lupus anticoagulant; RVVT, Russell Viper venom time; TT, thrombin time; VKA, vitamin K antagonist (eg, warfarin).

target value. The main disadvantage, common to all methods, is that it
does not permit 100% differentiation of “factor deficiencies” vs “fac-
tor inhibitors”. Another disadvantage is the additional need for labora-
tories to establish the cut-off value for discrimination of factor
deficiency vs inhibitor (or LA) and also to perform some additional ver-
ification. In general, the cut-off value will range from 10 to
15%.10,11,16,18,19
3.3 | Percent correction method
The “percent correction method”, sometimes also referred to as the
Chang method, after the method's advocate,20 reflects another com-
monly used method.10,11 The percent correction is identified by the
formula:
%correction = ½patient CT −1 : 1 mix CT Þ=ðpatient CT −NPP CT Þ x 100
where CT is again the clotting time of the test under investigation,
and NPP again reflects the normal pool plasma. Like the ICA, the cal-
culation is not complex and can be built into LIS/middleware/ instru-
mentation, and then an interpretation applied against a given target
value. The same main disadvantage, common to all methods, also
applies (it does not permit 100% differentiation of “factor deficien-
cies” vs “factor inhibitors”). Another disadvantage, like the ICA, is the
need to establish the cut-off value for discrimination of factor
deficiency vs inhibitor (or LA), and also to perform some additional
verification. In general, the cut-off value will range from 65 to
80%.10,11,20
3.4 | Other methods
There are of course many other potential options, especially for LA
investigation.21-24 As another example, a more subjective approach is
sometimes used. In the past, our laboratory used CT values of mix-
tures within 10 to 15% (depending on the initial prolongation value)
of the NPP result to identify “correction”, and any value above this to
reflect “non-correction”. However, such subjective approaches are
more difficult to standardize and incorporate into rules and algo-
rithms. For LA investigation, ratios of test results are instead some-
times used to identify the cut-off. For example, a ratio of 1.2 is
commonly applied to ratios to identify the absence of LA (value < 1.2)
or presence of LA (value ≥ 1.2).24 In essence, this is similar to using a
value of 20% for correction.
As a summary of this section, please refer to Table 2.
4 | L U P U S A N TI C OA G U LA N T S (L A ) A S A
SPECIAL CO N SIDERA TION
LA represent a laboratory representation of the clinical syndrome of
antiphospholipid syndrome (APS), and can also arise in other
FAVALORO 123

pathological states.27,28 For a more complete discussion of LA, the
reader is referred to an earlier test of the month publication in this
journal.24 Some mention of LA testing, mixing tests and alternate
approaches has already been mentioned above. It should also be
acknowledged that laboratory testing for LA is complex, with three
separate recent guidelines published.17,29,30 Of major relevance to the
current review is the inclusion of mixing test in all three guidance doc-
uments, albeit with different emphasis around importance or place-
ment in the LA diagnostic strategy in each publication.17,24,27-29,31
Workers in the field also differ in their opinions on whether confirma-
tory LA tests should be performed on neat (patient) test samples,
and/or on test mixtures (patient/normal plasma). In our laboratory, we
tend to perform mixtures on all patients under investigation for LA
when using the RVVT assay, but this is partly to offset or assess the
potential of anticoagulant therapy, which is commonly applied to
patients under “thrombophilia” investigation, otherwise leading to
false positives or false negatives.32,33 Accordingly, mixing tests in our
laboratory for LA investigation for RVVT represent the default
position, performed on both screen and confirmation tests, and we
then reflex to testing of neat patient samples for RVVT screen (and if
required RVVT confirm) on a case by case basis. In contrast, we tend
to perform APTT-based LA testing on non-mixture samples as the
default, and reflex to testing mixture samples, if required, on a case by
case basis. This combinational approach provides us with information
on likely presence/absence of LA vs potential for presence of antico-
agulants. Thus, the test patterns from individual cases can be assessed
and results will tend to point to one or other likelihood. Our labora-
tory may also perform additional reflex testing to evaluate these pos-
sibilities. Other laboratories would instead likely use different
approaches based on their personal experiences and potential biases.
As mentioned, different guidance documents vary in their recom-
mendations for testing of LA on patients on anticoagulant therapy,
with this also reflecting on mixing studies.17,24,27-29,31 The first guide-
line, from the ISTH SSC, for example, suggested LA testing was feasi-
ble for patients on VKA therapy for therapeutic INRs (international
normalized ratio), if testing was performed using mixing tests.29 Here,

F I G U R E 3 Potential different approaches to investigation of lupus anticoagulant (LA). (A) The classical approach, for example as proposed by
the ISTH SSC LA guidelines29 would promote use of mixing studies and place them early in the investigative pathway. (B) An alternative approach,
for example as considered in the CLSI LA guidelines,17 could position the mixing study later. (C) A third approach, in the context of integrated
testing, would even permit identification or exclusion of LA without performance of mixing studies (eg, as using the Werfen ACL Silica Clotting Time
(SCT) test method using SCT screen and SCT confirm).13 An alternative interpretation of integrated testing, as proposed by the CLSI LA guidelines,17
would instead incorporate a built-in mix method (eg, Diagnostica Stago StaClot LA- hexagonal phase assay). Since LA prolong clotting times by
inhibiting phospholipid, LA screen testing is done with low phospholipid concentration, to enhance detection of LA, and LA confirm testing is done
with high phospholipid concentration, to demonstrate that the clotting time shortens towards normal as the excess phospholipid overcomes LA
inhibition. Figure adapted from an educational slide set prepared by Gary Moore; used and adapted with author's permission. The original version is
available @ <https://diagnostics.roche.com/be/en/article-listing/coagyoulation.html>. APTT, activated partial thromboplastin time; DOACs, direct
oral anticoagulants; LR-APTT, lupus responsive-APTT; NPP, normal pool plasma; PL, phospholipid; NRI, normal reference interval; RI, reference
interval; RVVT, Russell Viper venom time; SCT, silica clotting time [Color figure can be viewed at wileyonlinelibrary.com]
124 FAVALORO

the normal plasma used will “replace” the missing (VKA affected) fac-
tors and normalize the test result accordingly, thereby increasing the
assay's specificity for LA.12,29 Thus, the default position for assess-
ment of VKA samples for LA would be performance of testing on mix-
tures, which is also essentially supported by the other LA
guidelines.17,30 In general, mixing studies on patients on heparin ther-
apy can help to dilute any heparin effect, and given that LA test
reagents tend to include heparin neutralizers, may help normalize test
results and eliminate any heparin effect for samples where heparin
levels are just above the therapeutic range in the neat patient sample.

TABLE 3 Anticipated patterns of routine test results and mixing tests for various scenariosa

Scenario
VKA therapy (start)/
vitamin K deficiency
(low level)
VKA therapy (stabilized
patient)/vitamin K
deficiency (high level)
Dabigatran
Rivaroxaban
Apixaban
Unfractionated heparin
(therapeutic level)
Unfractionated heparin
(above therapeutic level)
Factor VIII, IX, XI and/or
XII deficiency
Factor VII deficiency
Factor II, V and/or X
deficiency
Factor VIII, IX, XI and/or
XII inhibitor
Factor VII inhibitor
Factor II, V and/or X
inhibitor
Disseminated intravascular Prolonged
coagulation (DIC)
Liver disease
Additional testing/confirmation
PT test result APTT test result Mixing test result (if required)
Prolonged Normal or prolonged Both PT and APTT should correct TT will be normal if performed;
factors II, VII, IX, X will be low.
Prolonged Prolonged Both PT and APTT should correct TT will be normal if performed;
factors II, VII, IX, X will be low.
Normal or prolonged Prolonged APTT should not correct Dabigatran assay for confirmation.
(TT will be prolonged if
performed).
Prolonged Normal or prolonged PT should not correct Rivaroxaban (anti-Xa) assay for
confirmation. (TT will be normal
if performed).
Normal (may be Normal (may be If prolonged, PT/APTT should not Apixaban (anti-Xa) assay for
prolonged) prolonged) correct, but depends on method confirmation. (TT will be normal
employed to determine correction if performed).
(may correct into NRI)
Normal (unless Prolonged APTT should not correct Heparin (anti-Xa) assay; prolonged
reagent does not TT with non-correcting TT mix
contain heparin often suggestive, but may be
neutralizer, or level seen with other drugs such as
exceeds heparin dabigatran
neutralizer)
Prolonged Prolonged APTT should not correct; PT may or Heparin (anti-Xa) assay; prolonged
may not correct (may correct if TT with non-correcting TT mix
heparin diluted into range of often suggestive, but may be
heparin neutralizer) seen with other drugs such as
dabigatran
Normal Prolonged APTT should correct Factor VIII, IX, XI and/or XII assays
Prolonged Normal PT should correct Factor VII assay
Prolonged Prolonged/(normal) PT and APTT should correct Factor II, V and/or X assays
Normal Prolonged APTT should not correct (however, Factor VIII, IX, XI and/or XII
note that FVIII inhibitors are time (inhibitor) assays
and temperature dependent and
APTT may correct if
non-incubated mix performed)
Prolonged Normal PT should not correct Factor VII (inhibitor) assay
Prolonged Prolonged PT and APTT should not correct Factor II, V and/or X (inhibitor)
assays
Prolonged PT and APTT should correct D-dimer, fibrinogen, platelet
count, perhaps other factor
assays
Prolonged Prolonged PT and APTT should correct Factor assays, liver enzymes

Abbreviations: APTT, activated partial thromboplastin time; PT, prothrombin time; NRI, normal reference interval; TT, thrombin time; VKA, vitamin K
antagonist.
a
Some tests may or may not be prolonged with certain anticoagulants depending on specific test reagents used and concentration of drug.
FAVALORO
Mixing studies also act to dilute out DOAC related coagulation inhibi-
tion. On the other hand, mixing studies will also dilute out any LA
effect, and in some cases potentially lead to false negative LA find-
ings34; accordingly, any mixing study used in LA investigation needs
to balance these considerations.
As mentioned, the three most recent LA guidelines17,29,30 differ in
their placement of mixing studies in the LA diagnostic strategy. In the-
ory, mixing studies can also be omitted in the following circumstances:
(i) LA excluded because LA screening tests are normal - essentially
requires two tests, based on different principles (eg, APTT and RVVT
[Russell Viper venom time]) to be both normal; (ii) LA identified on
neat patient plasma by prolongation of screening test (containing low
level of phospholipid) with normalization of CT with confirmation
assay (containing high level of phospholipid), and thereby confirming
the phospholipid dependent CT prolongation (particularly supported
by the use of “integrated LA testing”). (Refer to Figure 3). In the ISTH
SSC guideline,29 in addition to suggesting mixing studies for VKA ther-
apy patients, mixing studies are also advocated as part of the strategy
to identify the inhibitory nature of LA,12 as originally also conceived
nearly three decades ago.35 Thus, mixtures of patient plasma and nor-
mal plasma should not correct if LA is present, since LA acts as a coag-
ulation inhibitor. However, occasionally, mixing studies falsely correct
despite LA, due to dilutional effects.34
All three LA guidelines, however, do concur on their recommenda-
tions to use a 1:1 mixture of patient:NPP for any mixing studies
performed,17,29-31 and to interpret test findings against an established
ICA or alternate mixing test-specific cutoff. Some examples of differ-
ent approaches to investigation of LA using mixing tests, including
placement, or not using mixing tests is shown in Figure 3.
5 | ANTIOCOAGULANTS - ADDITIONAL
CONSIDERATIONS
As already noted, mixing studies per se have limited utility for patients
known to be on anticoagulants, although some benefits may arise
should the anticoagulant in question not be known. Anticipated test
patterns for anticoagulants, both in terms of effects on common coag-
ulation tests (Table 3), as well as expected patterns on PT and APTT
including mixing tests (Table 3), can help guide patient follow up in
cases where the anticoagulation status is unknown. Thus, each antico-
agulant provides a unique “signature” in terms of test patterns36,37;
thereby, a patient sample giving a particular pattern of test results as
per Table 3 may accordingly be inferred to be on a particular anticoag-
ulant, and this can then be confirmed by specific testing.
6 | T I M E A N D TE M P E R A T U RE D E P E N D E N T
( E G , F V I I I ) I NH I B I T O R S - A S SP E C I A L
CONSIDERATION
As briefly mentioned earlier, mixing studies are usually performed as
an immediate mix test, meaning that patient plasma is mixed with
NPP and the test performed immediately thereafter. Most inhibitors
125
in hemostasis, including anticoagulants that inhibit coagulation fac-
tors, are immediately acting, and show type 1 kinetics.10,11 This means
that when test plasma containing such inhibitors are diluted with nor-
mal plasma in equal volume, about 50% of the inhibitory effect
remains evident, and so the CT of the test mixture will be about half
way between that of the test plasma and that of the NPP. Some inhib-
itors, however, notably those that develop against FVIII, are time and
temperature dependent,10,11,16 meaning that inhibitory effects are not
immediately evident, and only become fully expressed after some time
and at temperatures higher than ambient room temperature. Thus,
when immediate mixing tests are performed with such samples, the
mixtures may show a “false” correction, since the inhibitors are not
yet able to inhibit the factors from the NPP. If a FVIII (or other time
and temperature) inhibitor is present, or thought to be possible,
mixing studies would need incubation at 37°C (one to 2 hours is stan-
dard) in order to fully express the inhibitor, and to show a non-
correction in a mixture. This is similar to factor inhibitor studies using
a Bethesda or Nijmegen assay, where 2 hours incubation is standard.
In practice, this situation is not overly problematic, since clinical
patient history should always guide patient follow up. In addition,
results of reflex testing will also provide useful guidance. Thus, if the
patient is a known hemophilia patient, and thus a candidate for auto-
antibodies, or is a non-hemophilia patient but is suffering from bleed-
ing and is thus a candidate for allo-antibodies, then a timed incubation
of a patient/NPP mixture could be prepared and tested. In practice,
our specialized laboratory rarely ever does time-incubated routine test
mixing studies, instead finding it easier and more direct to undertake
factor level testing in these specific situations. If FVIII is low, then
consideration of FVIII inhibitors can then be made. However, timed
incubation mixing studies may be useful for those laboratories that
are not able to undertake factor assays. In addition, some laboratories
perform an incubated mixing study when factor VIII is low, to deter-
mine whether or not a Bethesda assay is needed to measure a factor
VIII inhibitor.
7 | C O N CL U S I O N S
In summary, mixing studies can aid laboratories in the investigation of
unexpectedly prolonged (“abnormal”) patient CTs, and help to differ-
entiate whether the “abnormality” reflects factor deficiencies vs inhib-
itors. Mixing studies are also useful for investigation of LA, although
guidelines differ in their view regarding utility and placement within
the strategy. There are advantages and limitations in all approaches.
No approach is 100% foolproof, and all will misidentify occasional
patients. On the other hand, all common approaches will correctly
identify most situations, and thereby help further guide laboratories
and clinicians to follow up testing. Table 3 and Figure 1 provide some
guidance on anticipated patterns and suggested follow up testing.
Also, just as many tests of routine coagulation may be identified to
have been unnecessarily ordered and potentially not followed up by
clinical staff, it should similarly be recognized that many mixing tests
performed by laboratories may ultimately also not lead to any evident
126
clinical follow up.38 However, some of those that are followed up will
inevitably lead to a reduction in patient morbidity as a result of appro-
priate action.26 Thus, a balance between performance of mixing stud-
ies on the one hand (time and costs) vs limitations and benefits may
need to be considered. Ultimately, mixing studies performed in view
of appropriate clinical scenarios (eg, bleeding patient) and for LA
investigations will have best utility, and the remainder can be consid-
ered as a kind of insurance policy that may or may not ever be
required (Summary Table). Also, all test processes, including mixing
studies, need to be verified or validated prior to implementation in the
laboratory.
DIS CLAIMER
The views expressed herein are those of the author and not necessar-
ily those of NSW Health Pathology. Naturally, workers in the field can
be identified to have a variety of perspectives, based on their own
experience.
ORCID
Emmanuel J. Favaloro https://orcid.org/0000-0002-2103-1661
RE FE R ENC E S
1. Bonar RA, Lippi G, Favaloro EJ. Overview of hemostasis and thrombo-
sis and contribution of laboratory testing to diagnosis and manage-
ment of hemostasis and thrombosis disorders. Methods Mol Biol.
2017;1646:3-27.
2. Lippi G, Franchini M, Favaloro EJ. Diagnostics of inherited bleeding
disorders of secondary hemostasis: an easy guide for routine clinical
laboratories. Semin Thromb Hemost. 2016 Jul;42(5):471-477.
3. Favaloro EJ. How to generate a more accurate laboratory-based inter-
national normalized ratio: solutions to obtaining or verifying the mean
normal prothrombin time and international sensitivity index. Semin
Thromb Hemost. 2019;45:10-21.
4. Favaloro EJ, Kershaw G, Mohammed S, Lippi G. How to optimize acti-
vated partial thromboplastin time (APTT) testing: solutions to esta-
blishing and verifying normal reference intervals and assessing aptt
reagents for sensitivity to heparin, lupus anticoagulant, and clotting
factors. Semin Thromb Hemost. 2019 Feb;45(1):22-35.
5. Favaloro EJ, Lippi G. Reference ranges in hemostasis testing: neces-
sary but imperfect. J Lab Precis Med. 2017;2:18.
6. Baluwala I, Favaloro EJ, Pasalic L. Therapeutic monitoring of
unfractionated heparin - trials and tribulations. Expert Rev Hematol.
2017 Jul;10(7):595-605.
7. Marlar RA, Clement B, Gausman J. Activated partial thromboplastin
time monitoring of unfractionated heparin therapy: issues and recom-
mendations. Semin Thromb Hemost. 2017;43(3):253-260.
8. Kottke-Marchant K. Algorithmic approaches to hemostasis testing.
Semin Thromb Hemost. 2014 Mar;40(2):195-204.
9. Mohammed S, Ule Priebbenow V, Pasalic L, Favaloro EJ. Develop-
ment and implementation of an expert rule set for automated reflex
testing and validation of routine coagulation tests in a large pathology
network. Int J Lab Hematol. 2019;41(5):642-649.
10. Kershaw G. Performance and Interpretation of Mixing Tests in Coag-
ulation. Methods Mol Biol. 2017;1646:85-90.
FAVALORO
11. Kershaw G, Orellana D. Mixing tests: diagnostic aides in the investiga-
tion of prolonged prothrombin times and activated partial thrombo-
plastin times. Semin Thromb Hemost. 2013;39(3):283-290.
12. Pengo V, Bison E, Banzato A, Zoppellaro G, Jose SP, Denas G. Lupus
anticoagulant testing: diluted russell viper venom time (dRVVT).
Methods Mol Biol. 2017;1646:169-176.
13. Tripodi A, Chantarangkul V. Lupus anticoagulant testing: activated
partial thromboplastin time (APTT) and silica clotting time (SCT).
Methods Mol Biol. 2017;1646:177-183.
14. Favaloro EJ, Bonar R, Duncan E, et al. Identification of factor inhibi-
tors by diagnostic haemostasis laboratories: a large multi-centre eval-
uation. Thromb Haemost. 2006;96:73-78.
15. Favaloro EJ, Bonar R, Duncan E. et al; On behalf of the RCPA QAP in
Haematology Haemostasis CommitteeMis-identification of factor
inhibitors by diagnostic haemostasis laboratories: recognition of pit-
falls and elucidation of strategies. A follow up to a large multicentre
evaluation. Pathology. 2007;39:504-511.
16. Detection KG. Measurement of Factor Inhibitors. Methods Mol Biol.
2017;1646:295-304.
17. Clinical and Laboratory Standards Institute (CLSI). Laboratory Testing
for the Lupus Anticoagulant; Approved Guideline. CLSI Document
H60-A. Wayne, PA: CLSI; 2014.
18. Rosner E, Pauzner R, Lusky A, Modan M, Many A. Detection and
quantitative evaluation of lupus circulating anticoagulant activity.
Thromb Haemost. 1987;57:144-147.
19. Benzon HT, Park M, McCarthy RJ, Kendall MC, Lindholm PF. Mixing
studies in patients with prolonged activated partial thromboplastin
time or prothrombin time. Anesth Analg. 2019;128(6):1089-1096.
20. Chang S, Tillema V, Scherr D. A “percent correction” formula for eval-
uation of mixing studies. Am J Clin Pathol. 2002;117:62-73.
21. Moore GW, Culhane AP, Daw CR, Noronha CP, Kumano O. Mixing
test specific cut-off is more sensitive at detecting lupus anticoagu-
lants than index of circulating anticoagulant. Thromb Res. 2016;139:
98-101.
22. Kumano O, Moore GW. Lupus anticoagulant mixing tests for multiple
reagents are more sensitive if interpreted with a mixing test-specific
cut-off than index of circulating anticoagulant. Res Pract Thromb
Haemostasis. 2017;2(1):105-113.
23. Kumano O, Moore GW. Ruling out lupus anticoagulants with mixing
test–specific cutoff assessment and the index of circulating anticoag-
ulant. Res Pract Thromb Haemostasis. 2019;3(4):695-703.
24. Favaloro EJ. The Russell Viper Venom Time (RVVT) test for investiga-
tion of lupus anticoagulant (LA). Am J Hematol. 2019;94(11):1290-
1296.
25. Ledford-Kraemer M. All mixed up about mixing studies. The Clotting
Times. 2004;3:2-6.
26. Blennerhassett R, Favaloro EJ, Pasalic L. Coagulation studies: achiev-
ing the right mix in a large laboratory network. Pathology. 2019;51(7):
718-722.
27. Favaloro EJ, Wong RCW. Antiphospholipid antibody testing for the
antiphospholipid syndrome: a synopsis of challenges and recent
guidelines. Pathology. 2014;46:481-495.
28. Miyakis S, Lockshin MD, Atsumi T, et al. International consensus
statement on an update of the classification criteria for definite
antiphospholipid syndrome (APS). J Thromb Haemost. 2006;4:
295-306.
29. Pengo V, Tripodi A, Reber G, et al. Update of the guidelines for lupus
anticoagulant detection. J Thromb Haemost. 2009;7:1737-1740.
30. Keeling D, Mackie I, Moore GW, Greer IA, Greaves M. British Com-
mittee for Standards in Haematology Guidelines on the investigation
and management of antiphospholipid syndrome. Br J Haematol. 2012;
157:47-58.
31. Moore GW. Recent guidelines and recommendations for laboratory
detection of lupus anticoagulants. Semin Thromb Hemost. 2014;40:
163-171.
FAVALORO
32. Favaloro EJ, Mohammed S, Curnow J, Pasalic L. Laboratory testing
for lupus anticoagulant (LA) in patients taking direct oral anticoagu-
lants (DOACs): potential for false positives and false negatives.
Pathology. 2019;51:292-300.
33. Favaloro EJ. Danger of false negative (exclusion) or false positive
(diagnosis) for “congenital thrombophilia” in the age of anticoagulants.
Clin Chem Lab Med. 2019;57:873-882.
34. Moore GW, Savidge GF. The dilution effect of equal volume mixing
studies compromises confirmation of inhibition by lupus anticoagu-
lants even when mixture specific reference ranges are applied.
Thromb Res. 2006;118(4):523-528.
35. Exner T, Triplett DA, Taberner D, Machin SJ. Guidelines for testing
and revised criteria for lupus anticoagulants. SSC Subcommittee for
the standardization of lupus anticoagulants. Thromb Haemost. 1991;
65(3):320-322.
36. Favaloro EJ, Lippi G. Laboratory testing in the era of direct or non-
vitamin K antagonist oral anticoagulants: a practical guide to
127
measuring their activity and avoiding diagnostic errors. Semin Thromb
Hemost. 2015;41:208-227.
37. Favaloro EJ, Pasalic L, Curnow J, Lippi G. Laboratory monitoring or
measurement of direct oral anticoagulants (DOACs): advantages, limi-
tations and future challenges. Curr Drug Metab. 2017;18(7):598-608.
38. Mohammad E, Thachil J. Mixing studies for abnormal coagulation screen
– the current trend. Clin Chem Lab Med. 2017;55(3):e54-e55.
How to cite this article: Favaloro EJ. Coagulation mixing
studies: Utility, algorithmic strategies and limitations for lupus
anticoagulant testing or follow up of abnormal coagulation
tests. Am J Hematol. 2020;95:117–128. https://doi.org/10.
1002/ajh.25669
128
T h e A m e r ic a n J o u r n a l O f H e m a t o l o g y
T E S T O F T H E M O N T H - SU M M A R Y T A B L E
COAGULATION MIXING STUDIES; UTILITY, ALGORITHMIC STRATEGIES AND LIMITATIONS FOR LUPUS
ANTICOAGULANT TESTING OR FOLLOW UP OF ABNORMAL COAGULATION TESTS
What is the test?
Mixing tests are indicated when an unexpected abnormal clotting time (CT) result is
obtained in a routine coagulation test, such as the prothrombin time (PT), or the acti-
vated partial thromboplastin time (APTT). The test performed as a mixing test is the
same test that originally gave the abnormal CT (i.e., PT, APTT, or both). A correction
in CT with the mixture suggests a factor deficiency or deficiencies (congenital or
acquired due to vitamin K deficiency, vitamin K antagonist [VKA] therapy [including
warfarin], disseminated intravascular coagulation [DIC], liver disease, etc). A non-
correction suggests a factor “inhibitor”, which may be a true factor “inhibitor” (auto-
or allo-antibody), or an anticoagulant acting as an “inhibitor” (eg, heparin, direct oral
anticoagulants [DOACs] such as dabigatran, rivaroxaban, apixaban), or a lupus anti-
coagulant (LA). Mixing tests can also be used for other routine tests such as throm-
bin time (TT) (for example, a prolonged TT that corrects is most likely due to
fibrinogen deficiency; which in turn can be congenital or acquired [e.g., DIC, clotted
sample]; in contrast, non-correction in a TT-mix may suggest heparin, argatroban,
bivalirudin or dabigatran). Mixing studies are also used as part of formal LA investiga-
tions, using APTT or RVVT (Russell Viper venom time).
How is the test measured?
Mixing tests are performed using the same clot-based tests that initially gave an
unexpected abnormal CT, be it PT, APTT and/or TT. For LA investigation,
mixing assays may be performed with the APTT and/or the RVVT. Accordingly,
mixing studies are normally reported in seconds, or as a ratio (eg, for LA tests),
or as an interpretation (eg, suggestive of factor deficiency, suggestive of factor
inhibitor, etc).
What are the normal range values?
The normal range values for mixing tests are typically the same as the routine
coagulation tests, or else some other methodology is applied to identify a cut-
off value to permit interpretation of a mixing test as being corrected or not
corrected. Examples of correction include mix test results within the normal ref-
erence range (or interval) as defined by the laboratory for that test, or alternate
laboratory validated methods that incorporate an index of circulation anticoagu-
lant (ICA) or percent correction. Sometimes assay ratios are used (eg, for LA
investigation), where a cut-off value of 1.2 is commonly reported as defining
absence of LA (<1.2) or presence of LA (≥1.2), as well as for correction of mixed
plasma tests.
What conditions or types of conditions is the test
used for?
Mixing tests are potentially useful to help investigate abnormal coagulation tests
and the need for further investigation to pinpoint the “defect” or reason for the
abnormal test result. In total, a wide variety of explanations can be proposed as
ultimately explaining the abnormal test result, and for which a mixing test may be
performed. Some examples include: factor deficiency or deficiencies (congenital or
acquired), vitamin K deficiency/antagonist therapy, DIC, liver disease, factor
“inhibitors”, heparin, DOACs, LA.
FAVALORO
What additional tests are helpful to do for a more
complete picture?
Mixing tests comprise one approach for investigation of an initially prolonged
clotting assay, especially PT and/or APTT. Other tests that may be useful depend on
the result of the mixing test(s). Note, TT testing is often useful, especially if the APTT
is initially prolonged. Factor assays may be suggested for performance if mixing tests
suggest factor deficiencies. Testing for DOACS or heparin may be useful if mixing
test patterns suggest this likelihood and anticoagulant status is unknown but clini-
cally important to determine. If LA is being investigated, then APTT and RVVT test-
ing may be performed, as may solid phase assays for antiphospholipid antibodies (eg,
anti-cardiolipin antibody (aCL) and anti-beta-2-glycoprotein-I (aβ2GPI)). Factor inhibi-
tor assays may be required if a factor inhibitor is suggested. Clinical background or
context is very important (patient history of thrombosis may point to an anticoagu-
lant; history of bleeding may point to a factor deficiency or inhibitor).
What tests provide similar information?
Mixing tests do not replace additional tests but help guide the direction towards
specific additional tests on a case by case basis. It is possible to omit performance
of mixing tests, but this may then require performance of a larger number of
“diagnostic” assays to pinpoint the reason for any initial test prolongation. For
example, a prolonged APTT that mixes to normal can usually avoid the need to
investigate the presence of “inhibitors” or DOACs. If mixing tests are not per-
formed, reflex testing may require extensive testing to investigate all possible
scenarios.
What tests can be avoided/eliminated?
As per above, mixing tests that correct can guide laboratories to not undertake
testing for inhibitors/DOACs. However, mixing tests that do not correct can
guide laboratories to not undertake testing for factor levels if DOACs are instead
suggested by the pattern of test results.
How does its use impact treatment?
The results of mixing tests may or may not impact treatment, depending on the
reason for the originally identified prolonged CT, the results of the mixing tests,
and the clinical context. As an example, a mix test result that corrects and which
is reflected by a clinical scenario of bleeding may suggest replacement factor
therapy may be indicated or effective (eg, fresh frozen plasma in the case of
trauma induced coagulopathy; specific factor replacement if FVIII deficient, etc).
As another example, a mix test result that does not correct and which is
reflected by a clinical scenario of bleeding may reflect a factor inhibitor, help
lead to confirmation of same, and then guide treatment with either factor
replacement or alternate therapy dependent on the level of inhibitor eventually
identified. As a final example, a mix test result that does not correct and which
is reflected by a clinical scenario of thrombosis or pregnancy morbidity may ulti-
mately identify a patient with antiphospholipid syndrome, and thus guide spe-
cific therapy for that condition.