Control of Anticoagulation Subcommittee

19th July 2002
9:00-13:00
Imperial Room
Boston Park Plaza Hotel

Chairman:  M. Greaves, UK
Co-chairs:  H. Bounameaux, Switzerland; J. Harenberg, Germany; C. Kearon, Canada; F. R. Rosendaal, The Netherlands;
S. Schulman, Sweden; A. Tripodi, Italy; A.M.H.P. van den Besselaar, The Netherlands

Working Party on Duration of Anticoagulant Therapy in Venous Thromboembolism
Chair: C Kearon [S Schulman assisted in the design of the programme]

Introduction   C Kearon
Dr. Kearon introduced the session by reviewing the current evidence base for duration of anticoagulation after a first VTE:
Influence of presentation (DVT or PE) on risk of recurrence
Dr  J Douketis reviewed the evidence for difference in recurrence rates after initial DVT or PE [submassive].  A meta-analysis of 25 studies  [JAMA 1998, 279:458] and some additional reports were cited.  Overall 3 month recurrence after first PE was not significantly different from that after first DVT, at around 5% for both.  However the meta-analysis indicates that the impact of recurrence after initial PE is greater with an approximately three-fold greater mortality.  At least some of this is attributable to co-morbidity, especially cardiorespiratory disease.  There is also evidence for an approximately two-fold higher recurrence rate after first iliofemoral DVT in comparison to femoral or popliteal DVT.

Dr J Heit reported on the Mayo Clinic VTE study which is based on the Rochester Epidemiology Study. An overall recurrence rate of 30% at 10  years was reported, being highest early.  Independent predictors of recurrence were greater age, higher BMI, malignancy and neurological disease.  Lower recurrence rates were found when the presenting VTE was associated with oral contraceptive or HRT use or gynaecological surgery, supporting other reports of lower recurrence rates in subjects with transient risk factors.

Dr R White reported on his study based on linked hospital discharge data in California [In press, Thrombosis and Haemostasis].  Similar rates of recurrence to those previously reported were found and a higher prevalence of cardiorespiratory disease was noted among patients with PE compared to those with DVT.  It was confirmed that those presenting with DVT were much more likely to manifest DVT as the recurrent event whereas those with PE were more prone to PE recurrence [60-80% of patients with recurrent VTE as PE had index PE].  An analysis of time to second event did not support diagnostic bias as a cause of this phenomenon.
Does extending duration of treatment reduce subsequent cancer incidence?   
Dr G Agnelli reported on the WODIT study.  This addressed the question whether extending the duration of anticoagulant treatment after first VTE from 3 months to one year reduces the chance of a new diagnosis of cancer.  In a total of 429 cases randomised to one or other group there was no significant difference in number of new cancer diagnoses between the two arms.  This study provides no evidence for a reduction of cancer risk with oral anticoagulant therapy.

Can markers of thrombosis after stopping anticoagulant therapy predict recurrent VTE?   
Dr G Palareti referred to 3 early studies which suggested a rise in plasma D-Dimer concentration after discontinuing oral anticoagulation in some subjects treated for VTE.  He went on to review his recently published study which was designed to expand on these observations.  In a cohort of around 400 patients, D-Dimer measured after discontinuing oral anticoagulant was significantly higher in those subsequently suffering VTE recurrence.  The negative predictive value of normal D-Dimer after 3 months of treatment was 96%. Subgroup analysis suggested that this was also the case in subjects with identified heritable thrombophilia.  The higher D-Dimer levels did not seem to relate to poorer recanalisation, although the highest recurrence rates were in the subgroup with no recanalisation and raised D-Dimer concentration.   It was noted that Faltorini et al [Thromb. Haemost. 2002; 88:162] have also recently reported a high negative predictive value for D-Dimer measured during anticoagulation.
Are residual venous abnormalities predictive of recurrent VTE?
Dr C Kearon and Dr F Piovella reviewed data on residual vessel occlusion and recurrence risk.  In the DURAC study it was found that there was a higher recurrence risk in the contralateral than the ipsilateral leg on long term follow-up.  This does not suggest importance for local factors in the originally affected veins.  In two small studies there was no evidence for significantly higher recurrence risk associated with persistence of abnormal ultrasound or plethysmography findings; however, in a more recent study by Prandoni et al normalisation of ultrasound findings was associated with lower risk of recurrence.  Dr Piovella reviewed his recently published study [Haematologica 2002, 87:515] in which normalisation of ultrasound findings predicted lower risk of recurrence.  As a result a study in which ultrasound findings and D-Dimer assay are combined is about to commence.
Is factor V Leiden more of a risk factor for DVT than PE?   
Dr M Prins reported on evidence that heterozygosity for factor V Leiden carries a relatively higher risk of DVT than of PE.  He noted that several studies have been remarkably consistent in demonstrating an around 2-fold higher risk of DVT than of PE in carriers when compared with VTE patients without V Leiden.  In one study there was no such effect of heterozygosity for G20210A.  It was postulated that V Leiden may be associated with less friable clots and/or more local inflammation.
Outcome from the Working Group.
Dr Kearon summarised these presentations and this allowed the following clinically useful conclusions to be drawn:
Working Party on Near-patient Testing and Self-management of Oral Anticoagulant Control and the ECAA Steering Group
Chair:    AMHP van den Besselaar/A Tripodi

The ECAA multicentre calibration of the CoaguChek Mini and TAS PT-NC monitor systems
A Tripodi
Reference was made to 2 published papers assessing the possibility of switching to use of whole blood for calibration and to assess simplified calibration using recalcified plasma in place of whole blood.  The CoaguChek Mini and TAS analysers were assessed.  It was concluded:
  1. The proposed modified ISI calibration method is reliable for both monitors.
  2. For both monitors the CV of the calibration for both whole blood and plasma is lower than 5% but higher than the recommended 3%.
  3. Plasma may be suitable for determination of ISI in relation to both monitors.
The use of ECAA lyophilised plasmas for calibration of the CoaguChek Mini and TAS PT-NC monitor systems
Dr A van den Besselaar
Results of experiments with 3 types of plasmas were reported:
Artificially depleted [n=60]
Coumarin plasmas [n=60]
Normal [n=20]

Two monitors were assessed but not identified in the presentation.  With one lyophilised coumarin samples ISI agreed with whole blood ISI, with the other there were discrepancies.
On attempting to simplify procedures by reducing the number of lyophilised plasmas for ISI calculation, it was found that a minimum of 30 coumarin and 10 normals sufficed for one machine and 20 and 7 for the second.

The ECAA system for quality control of the CoaguChek Mini and TAS PT-NC monitor systems
Dr Michelle Keown
The need for methods to facilitate the QC of individual monitors was stressed.  An ECAA study at 10 national centres has been performed on TAS and CoaguChek monitors.  Sixty artificially depleted plasmas and 60 coumarin plasmas were employed.  Five and 3 of these, respectively, were selected which gave good INR distribution in both systems and CV <15%.  There were considerable INR differences between centres on the full plasma set and reflected in the selected smaller set.  There was little difference in results between sets of 3 and 5 plasmas.  There was no clear advantage in use of coumarin plasmas over depleted plasmas.  Examples of the use of plasmas for individual QC were given and it was concluded that provision of sets of 3 ECAA plasmas is a simple and practical method for checking reliability of INR measurements on individual monitors of both types.

Overview of the results of the ECAA Normalisation and Standardisation of Home Monitors PC Study
Dr L Poller
An overview of the group’s publications on this topic was presented.  Additional work in press was reviewed.  This has demonstrated significant instrument variability over 14  monitors  [both TAS and CoaguChek].  There was obvious greater dispersion of results with one monitor type than the other [types not identified].  In a second study equivalence in quality was shown between NPT and laboratory testing with computer assisted dosing employing a randomised cross-over design.  Finally the reliability of the displayed INR on the two monitors has been assessed in 10 centres using six hundred coumarin patients’ whole blood and plasma IRP manual PT tests.  Displayed INR averaged around 20% higher on one system than the other.  This was accounted for by consistent overestimation of the ‘true’ INR on one system and consistent underestimation on the other.  It is concluded that extra steps in ISI calibration and/or QC described in the ECAA Technology Implementation Plan are needed.

Outcome from the Working Group:
The ECAA Technology Implementation Plan incorporating ISI calibration and QC of POCT monitors has been accepted by the European Commission.

It was noted that the activities of the group had spanned many years and provided invaluable data.  Although it is accepted that more work is needed it was agreed that an SSC report should be prepared at this stage in order to provide state of the art information to interested parties.  Professor Preston agreed to liaise with the other members in the drafting of a document with a deadline of January 2003.

Influence of sample optical clarity on INR determinations on fresh, frozen and lyophilised test samples
Dr S Kitchen
A study of the effects of sample optical clarity on INR determination was reported.  INR was determined using Innovin/Sysmex CA 6000, PT Rec and PT Fib HS S Plus, both on ACL. Local MNPT and verified manufacturers’ instrument specific ISI were used.  Fresh, thawed frozen and reconstituted lyophilised plasmas were examined.  Opacity was determined by light scatter and expressed as an ‘h’ value.  A significant proportion of samples from subjects on warfarin were found to have a high h value [2% >70].  The study clearly demonstrated that freezing and freeze drying increase opacity and this can raise the measured INR with  PT Rec but not with the other 2 reagents tested.  Such an effect is seen in fresh samples on occasions.  These observations are highly relevant in proficiency testing and should be noted by scheme managers.  There may be implications for INR calibration plasmas also.

Status of international reference preparations for thromboplastin
Dr A van den Besselaar/Dr A Tripodi
The MNPT of rTF/95 and RBT/90 between 1995 and 2000 was noted to be stable.  The residual stocks of  these reference preparations were illustrated, indicating that, based on previous usage rates, there should be sufficient rTF95 for over 5 years but only 3 years for RBT/90.  It was concluded that plans should be made and implemented for the replacement of RBT/90 and proposals were presented for the mechanism to achieve this under the auspices of the SSC.
Outcome:
The proposals were accepted by the Subcommittee and are to be implemented forthwith.

A new algorithm for computer-assisted dosing: Presentation and demonstration
Dr F R Rosendaal
The need for more accurate dosing was identified, especially as bleeding risk is increased with greater fluctuation in the INR in a treated individual as well as through overtreatment.  A new model was described which incorporates an estimate of the subject’s previous dose sensitivity.  This has been applied to a dataset of over 200 and gave a small predictive error.  The algorithm is to be tested prospectively in a randomised study of 500 patients with time in therapeutic target range as the primary endpoint and clinical events as a secondary endpoint.
Outcome:
Dr Rosendaal indicated that the system would be made available to interested parties should the potential advantages be confirmed in the prospective study.

WHO Working Group on Unfractionated Heparin and Low Molecular Weight Heparin / Working Party on Monitoring of Low Molecular Weight Heparin
Chair:    T Barrowcliffe/M Greaves

Progress on global harmonisation of heparin assays
Drs A Padilla/E Gray
There was nothing new to report

Progress on the 2nd WHO standard for LMW heparin
Dr Elaine Gray
The pilot study was reviewed.  Anti-Xa and anti-IIa chromogenic assays were employed in 11 laboratories in the study.  It had been concluded that UFH should not be used as a calibrant for LMWH due to poor interlaboratory agreement and nonparallelism.  All LMWH were comparable in the study, but 2 samples tended to non-parallelism.  The study allowed selection of two samples with the lowest interlaboratory variation to be selected for further study.

The protocol for this new study was described.  The two candidates [above] are to be compared with the 1st International Standard.  Participants will be recruited in July/August 2002.  It was noted that there is a need for volunteer clinical laboratories, and those interested should contact Dr Gray.
There will be a report of the findings to the SSC in 2003 and a replacement standard will be recommended.

Status of Paediatric Guidelines
Dr. P Massicotte
It is considered that evidence-based guidelines cannot yet be formulated.  There is a need for more safety, efficacy and monitoring studies.

Anti-Xa levels during prolonged therapy with LMW heparin in cancer
Dr M Kovacs
A study was designed to determine whether accumulation of the anticoagulant effect of LMWH occurs when administered long term in patients with cancer.  Twenty-four patients from a single centre were evaluated.  The results were reassuring in that anti-Xa levels measured at 4 to 6 hours after injection were comparable at days 7 and 28 of therapy [1.11 and 1.03 respectively, p 0.37]

Problems in the evaluation of anticoagulant activity of new direct and indirect factor Xa inhibitors
Dr M Samama
The need for accurate monitoring of some patients treated with these new agents was acknowledged.  A study of assays of Fondaparinux [pentasaccharide] and DX9065a [a synthetic propanoic acid derivative] was made. The questions of preferred anti-Xa method, most appropriate expression of results  [activity or gravimetric] and heterogeneity of results in global tests were addressed.  It was concluded that expression of anti-Xa on a gravimetric basis has some advantages and standardisation of anti-Xa measurement for these new drugs is required.  Also the heterogeneous response in more global tests was confirmed and requires further study and explanation.  Work is ongoing on these issues.

CTAD (Citrate, Theophylline, Adenosine, Dipyridamole) as a superior anticoagulant to citrate for heparin monitoring
 Dr S Kitchen
A study was performed to examine the practical significance of previous reports that CTAD [citrate, theophylline, adenosine, dipyridamole ]  is useful in the prevention of loss of heparin over time in clinical samples, through platelet stabilisation and limitation of PF4 release.  Samples were taken into citrate or CTAD and analysed immediately and after mixing for  4 hours.   PT/INR was assessed using Innovin and PT Fib HS Plus, APTT with Actin FS and anti-Xa by chromogenic assay.  Samples included 32 normals, 32 warfarinised, 30 on UFH and 30 on LMWH.  A 32%  loss of anti-Xa activity with UFH was noted in citrate, with around 25% shortening of APTT.  This was abolished by use of CTAD.  Samples in CTAD were suitable for LMWH treated subjects.  There were some differences in results with PT reagents on CTAD samples from warfarinised subjects, and INR values were slightly higher using either reagent in CTAD compared with citrate.  It is concluded that CTAD is the anticoagulant of choice for UFH monitoring.  For general use in the coagulation laboratory for anticoagulant monitoring, it is not yet clear whether the heparin protection effect is sufficiently advantageous to overshadow the small effect of CTAD on INR.


Working Party on Standardisation of Methods to Determine Direct Thrombin Inhibitors
Chair: M Greaves

Results of the International Interlaboratory Thrombin Inhibitor Study
Dr Elaine Gray
The report of the collaborative study of assay methods for hirudin and argatroban was presented.  Spiked plasmas were examined using APTT [local and common reagent~Actin FS, 10 laboratories], Anti-IIa chromogenic assay, Ecarin clotting time by wet and dry chemistry methods, ELISA for hirudin.  Four(4) sets of plasma samples were examined in each method.  In summary, the APTT and ECT methods gave low inter-laboratory variability [%GCV under 5 for most labs].  There was greater variability using anti-IIa and hirudin ELISA.  In interlaboratory analysis common APTT and dry ECT performed well.  There was relatively higher variability on hirudin samples with local APTT.  In the wet ECT and anti-IIa assays, the variability increased with concentration of analyte.  The hirudin ELISA performed poorly.  In terms of sensitivity all methods gave linear dose-response and were able to detect 0.31µgm/ml.  Overall, although APTT gave good reproducibility within and between labs, the sensitivity was lower than the other methods.  
Outcome:
There will be a written report to participants in August 2002 and a report to subcommittee members in October/ November 2002.  An SSC communication will be submitted for publication early in 2003.
It is acknowledged that further studies are needed, especially those on ex vivo samples from treated patients.  Possible collaboration in these further studies with relevant pharmaceutical companies was raised.

Working Group on Calibrated Plasmas for INR Determination
Chair: T Barrowcliffe

NCCLS Guideline on procedures for validation and calibration of the International Normalised Ratio using the Prothrombin Time test.
Dr Dorothy Adcock
A brief update on the NCCLS Subcommittee on PT Calibration and the forthcoming guideline was presented.  The major points and recommendations were reviewed.  The group were congratulated on the progress achieved.

Discussion of final report of the SSC Working Group on Calibrated Plasmas for INR Determination
Dr T Barrowcliffe

Outcome from the working group:
After detailed discussion of the problems, a vote was taken on the need for preparation and dissemination of a guideline on this topic rather than the review manuscript which had been proposed as a solution to the lack of consensus among the working group.  The result was in favour of a guideline in  which areas of lack of consensus would be acknowledged.  It was agreed that Dr van den Besselaar would take over Chairmanship of the working group and undertake drafting of the guideline alongside the other members.  It was suggested that Dr Greaves could assist as a non-cognate facilitator in this process.