One sample, several different answers — D-dimer is one of the least standardised assays in coagulation.
D-dimer measures a fibrin degradation fragment released when plasmin lyses a cross-linked clot — the read-out that drives venous thromboembolism exclusion. The problem is the number itself: the same plasma reported in fibrinogen-equivalent units (FEU) reads roughly twice what it reads in D-dimer units (DDU), and across assays a single sample can vary up to twenty-fold. Get the unit, the cut-off or the age adjustment wrong and a safe exclusion becomes an unsafe one. Technoclone supplies the TECHNOLEIA D-Dimer Latex Kit, Technochrom Plasminogen and Technochrom Plasmin Inhibitor, plus the research-use Technozym Glu-Plasminogen ELISA — we advise on how the numbers actually compare.
The same plasma, divergent numbers — units and assays pull apart.
There is no shared scale across D-dimer assays. A result reported in FEU is around double the same result in DDU; choose two different manufacturers' assays and the spread widens further. The chart below shows one sample read across units and methods, with the fibrinogen-equivalent-to-D-dimer-unit factor of roughly two.
The chart shows one sample reported across assays and units — solid bars are FEU-reported, outlined bars DDU-reported, against the 500 µg/L FEU cut-off. In numbers: 1 µg/mL FEU ≈ 0.5 µg/mL DDU — fibrinogen-equivalent units read roughly 2× the equivalent D-dimer-unit value. 28 different unit combinations are in use; the ISTH SSC recommends a single unit, FEU (µg/L or mg/L). Identical samples can vary up to 20-fold between assays, with inter-laboratory CV of the order of 23–25% (Olson 2013), with variability greatest near the 500 µg/L FEU cut-off.
Why there is no clean head-to-head — and what does hold across assays.
D-dimer is unusual: the assays are not comparable by design, so a head-to-head correlation between products is not the meaningful axis. The defensible figures are the harmonisation facts — the unit relationship, the scale of variation, and the validated age-adjustment rule — together with the reference ranges for the fibrinolytic markers around them.
Fibrinogen-equivalent units read about twice the value of D-dimer units for the same sample — 1 µg/mL FEU ≈ 0.5 µg/mL DDU. With 28 different unit combinations in circulation, the ISTH SSC recommends standardising on a single unit: FEU (µg/L or mg/L). The factor of two is approximate, and it does not let you convert a value reported by a DDU assay into a comparable FEU number.
The factor of two describes the unit relationship, not an interchangeable result: it is approximate and cannot be used to convert a DDU-reported value into a comparable FEU number. There is no clean head-to-head by design — values are not comparable across assays.
Run an identical sample through different assays and the result can vary up to twenty-fold. Inter-laboratory CV runs of the order of 23–25% (Olson 2013), and variability is worst close to the conventional 500 µg/L FEU clinical cut-off — exactly where a misread changes a decision. This is why the unit and the assay must be stated, not assumed.
The 20-fold spread is the extreme seen across method-diverse assays, and the ~23–25% inter-laboratory CV (Olson 2013) reflects that same method diversity — neither is a fixed property of any single assay. They make the point that the unit and the assay must be stated, not assumed — there is no clean head-to-head by design.
Age-adjusted cut-off — age × 10 µg/L FEU in patients ≥50 years
For patients aged 50 and over, an age-adjusted cut-off of age × 10 µg/L FEU safely raises the proportion in whom pulmonary embolism can be excluded without imaging. In the ADJUST-PE study (n=3,346), exclusion in patients aged 75 and over rose from 6.4% to 29.7%, with a three-month thromboembolic failure rate of just 0.3%. The rule is defined in FEU, which is precisely why the reported unit matters at the point of decision.
The age-adjusted rule is defined in FEU and was validated on the specific assays used in the trial; the published rule should not be transferred to a DDU-reported value without a separately derived DDU threshold and assay-specific validation, and confirmation against the local assay is needed before adoption. This is why the reported unit matters at the point of decision.
Plasminogen and alpha-2-antiplasmin — the enzyme and its brake
D-dimer is the product; the system that makes it is plasmin, generated from plasminogen and held in check by alpha-2-antiplasmin. The chromogenic plasminogen assay reads a normal range of roughly 80–120% (illustrative, lab-dependent — per assay IFU); a low result points to congenital plasminogen deficiency, the cause of ligneous conjunctivitis. Alpha-2-antiplasmin is the primary inhibitor of plasmin, and its deficiency is a rare bleeding diathesis.
The 80–120% plasminogen range is illustrative and lab-dependent — confirm against the assay IFU. Antiplasmin reference values are generic; a Technochrom Plasmin Inhibitor percentage is not quoted here without the IFU. The ~2× factor is approximate and does not apply to DDU assays as a conversion.
D-dimer, plasminogen and antiplasmin — with the units spelled out.
For an assay this poorly standardised, the work is as much about the reporting as the reagent. We advise on the panel and on the units, cut-offs and age-adjustment your laboratory reports against — so a number means the same thing to the clinician reading it.
The fix is harmonisation, not a better number — standardise on one outcome-validated assay across the network.
If a single sample can read twenty-fold apart between assays, no individual D-dimer is the answer; the answer is to stop mixing them. The practical resolution is to fix one D-dimer across every site in the network — an assay that is outcome-validated for VTE exclusion, supports the age-adjusted cut-off (age × 10 µg/L FEU), and is reported in FEU — and then report it the same way everywhere, so a result moved between sites still means the same thing.
elixir's value sits in that selection and validation work — comparing candidate assays method-fairly against your case mix, confirming the age-adjusted rule holds on the chosen assay before adoption, and setting the unit and reporting convention across the network. The right choice is frequently a competitor's IVD, and we will say so. Where the chosen line is one we represent for Technoclone, we disclose that interest; where it is not, the advice stands unchanged. There is no independent head-to-head between D-dimer assays by design, so this is judgement applied to your data, not a league table.
The Technoclone panel
- TECHNOLEIA D-Dimer Latex Kit — a latex-enhanced immunoassay for the D-dimer fibrin-degradation marker; front-line VTE-exclusion use turns on local outcome-validation (see the harmonisation note above).
- Technochrom Plasminogen — chromogenic plasminogen activity, normal range ~80–120%.
- Technochrom Plasmin Inhibitor (alpha-2-antiplasmin) — the primary plasmin inhibitor; deficiency is a rare bleeding diathesis.
- Technozym Glu-Plasminogen ELISA — antigen measurement of native Glu-plasminogen (research use only).
What elixir does
As UK partner for Technoclone, we advise on the harmonisation that surrounds these assays — which assay to standardise on, which unit to report, where the age-adjusted cut-off applies, and why a value from one assay cannot be read against another. The selection advice is method-fair, and may land on a competitor's IVD; we disclose where a recommended line is one we represent for Technoclone.
Any modelling or computational work we offer around assay design or unit reconciliation is a human-led service — scoped, run and signed off by a person — not a product on the shelf.
The figures on this page, and where they come from.
- Thachil J, et al. ISTH SSC — FEU ≈ 2× DDU (1 µg/mL FEU ≈ 0.5 µg/mL DDU); 28 unit combinations in use; recommends FEU (µg/L or mg/L). JTH 2020;18(9):2408. PMC7307061
- Spannagl M, Haverkate F, Reinauer H, Meijer P. Identical samples vary up to 20-fold between assays, with high variability around the clinical cut-off. Blood Coagul Fibrinolysis 2005;16(6):439-43. PMID 16093735
- Olson JD, Cunningham MT, Higgins RA, Eby CS, Brandt JT. D-dimer: simple test, tough problems — inter-laboratory CV 22.8–25.5%; method-specific means varied 7.1–21.6-fold. Arch Pathol Lab Med 2013;137(8):1030-8. PMID 23899057
- Righini M, et al. ADJUST-PE — age-adjusted cut-off age × 10 µg/L FEU (≥50y); PE exclusion in ≥75y rose 6.4% → 29.7%; 3-month failure 0.3% (n=3,346). JAMA 2014;311(11):1117. PMID 24643601
Pricing a D-dimer or fibrinolysis panel, or untangling the units you report?
Tell us which markers you need and the units your laboratory reports against. We will reply with how we would approach the panel, what it would involve, and an honest quote — usually within two working days.