Reading the 4Ts Score for Heparin-Induced Thrombocytopenia

Heparin-induced thrombocytopenia, or HIT, is an immune drug reaction in which antibodies against platelet factor 4-heparin complexes activate platelets and create a dangerous prothrombotic state: the platelet count falls, but clotting rather than bleeding is the defining clinical threat. Because many hospitalized patients receive heparin and have other reasons for thrombocytopenia, indiscriminate antibody testing produces false alarms. The 4Ts score estimates probability before testing.

Why pretest probability matters#

PF4-heparin immunoassays are sensitive but can detect antibodies that do not activate platelets. Positive antibodies are common after some surgeries and do not always mean clinical HIT, and if a low-risk person is tested, a positive result may be false and can trigger costly non-heparin anticoagulation, bleeding, delayed procedures, and an inaccurate lifelong allergy label.

Conversely, delayed recognition of true HIT permits thrombosis. The score is therefore part of a linked diagnostic and management pathway, not paperwork you fill in after ordering the test. ASH recommends the 4Ts rather than unstructured gestalt for suspected HIT; it recommends against laboratory testing and empiric HIT treatment when the score is low, unless score uncertainty such as missing data warrants caution.

T1: thrombocytopenia magnitude#

This domain uses both relative fall and lowest count:

• 2 points: platelet fall greater than 50%, with nadir at least 20 × 10^9/L.
• 1 point: fall 30% to 50%, or nadir 10 to 19 × 10^9/L.
• 0 points: fall less than 30%, or nadir below 10 × 10^9/L.

The low-nadir rule may look backwards the first time you use it. HIT often produces moderate thrombocytopenia; an extremely low count is more typical of another process, although exceptions and overlapping causes occur.

Use the highest appropriate platelet count after heparin starts or the recent baseline, depending on the clinical course. A fall from 350 to 160 is greater than 50% even though 160 may be reported normal. A fall from 90 to 55 is less than 50% and scores differently; platelet transfusion, surgery, chemotherapy, extracorporeal circuits, infection, and missing measurements can obscure the true peak and nadir. Show the plotted count trajectory rather than relying on two selected values.

T2: timing of the fall#

Typical immune HIT begins five to ten days after the first recent heparin course because antibodies need time to develop. The standard timing points are:

• 2 points: clear onset on days 5 to 10, or a fall within one day when heparin was received in the previous 30 days.
• 1 point: timing compatible with days 5 to 10 but unclear because counts are missing; onset after day 10; or a fall within one day with heparin in the previous 30 to 100 days.
• 0 points: fall within four days without recent heparin.

The first day of heparin is day 0 in common scoring instructions. Every source counts: unfractionated or low-molecular-weight heparin, flushes, catheters, dialysis, procedural doses, and recent admissions. A medication administration record may be more reliable than a single active-medicine list.

Postoperative platelet counts often fall immediately through dilution and consumption, recover, and then fall again around days 5 to 10, and this biphasic pattern after cardiac surgery is more suspicious than an immediate fall that remains low. Timing must be interpreted against the expected clinical trajectory. Delayed-onset HIT can present after heparin has stopped, and spontaneous HIT-like syndromes are rare. These require specialist reasoning beyond a mechanical score.

T3: thrombosis and related manifestations#

The third domain scores new or progressive thrombosis and characteristic reactions:

• 2 points: new confirmed thrombosis; skin necrosis at heparin injection sites; or an acute systemic reaction after an intravenous unfractionated-heparin bolus.
• 1 point: progressive or recurrent thrombosis; erythematous non-necrotizing skin lesions; or suspected thrombosis not yet confirmed.
• 0 points: none.

HIT can cause venous or arterial thrombosis, limb ischemia, pulmonary embolism, stroke, myocardial infarction, adrenal injury, or circuit and catheter clotting. A platelet fall may precede thrombosis, so absence does not rule it out.

Do not double-count a single event. A suspected clot that is later excluded should be rescored. Conversely, a new confirmed event found after the initial calculation can raise probability and change the pathway.

T4: other causes of thrombocytopenia#

This is often the least reproducible domain:

• 2 points: no other apparent cause.
• 1 point: another cause is possible.
• 0 points: another cause is definite.

Possible alternatives include sepsis, disseminated intravascular coagulation, major surgery, extracorporeal support, intra-aortic balloon pump, massive transfusion, medications, chemotherapy, liver disease, immune thrombocytopenia, thrombotic microangiopathy, marrow failure, and pseudothrombocytopenia.

The question is not whether any alternative exists. It is how convincingly it explains the magnitude and timing. Sepsis and HIT can coexist. Calling sepsis “definite” without evaluating the platelet pattern can under-score HIT; assigning 2 points whenever uncertainty exists can over-score it. Document the alternatives and the evidence for each. A reproducible score says why you gave this domain 0, 1, or 2.

Total score and validated performance#

Add the four domains: 0 to 3 low probability, 4 to 5 intermediate, and 6 to 8 high. In the 2012 systematic review and meta-analysis, a low score had a negative predictive value of 0.998, 95% confidence interval 0.970 to 1.000, which makes the tool particularly useful for ruling out HIT.

Positive predictive value was much lower for intermediate scores and imperfect even for high scores. Prevalence, population, assessor, and reference standard affect performance. The category is not a post-test probability that can replace laboratory evidence, and implementation studies show that scoring errors and interobserver disagreement occur, especially for timing and alternative causes, and an electronic calculator cannot fix an input you got wrong.

What happens after a non-low score#

ASH's pathway recommends a PF4-heparin immunoassay for intermediate or high clinical probability, and when positive and a functional assay is available, a platelet-activation test such as the serotonin-release assay is generally used to improve specificity.

Immunoassay strength matters. A strongly positive ELISA optical density combined with a high 4Ts score raises likelihood more than a weak positive with an intermediate score. Cutoffs and assay types differ. Rapid immunoassays and IgG-specific assays do not have identical performance.

Functional assays detect whether patient antibodies activate donor platelets under test conditions. They are more specific but technically demanding, may have delayed turnaround, and can rarely be falsely negative. Results must be integrated with clinical probability and immunoassay strength.

Management while tests are pending depends on the score, bleeding risk, indication for anticoagulation, kidney and liver function, clinical stability, and local assay timing. ASH provides distinct pathways for intermediate probability with high bleeding risk and for high probability. This is urgent clinician-directed care; a patient or nonspecialist should not start, stop, or substitute anticoagulation from an online calculation.

Common scoring errors#

Using only the absolute count: HIT is defined by the fall pattern as well as nadir.

Starting the clock at hospital admission: recent procedures, flushes, and prior admissions may contain the relevant heparin start.

Calling every early postoperative fall HIT: expected immediate postoperative thrombocytopenia has a different pattern; a second fall later is more concerning.

Ordering the antibody first: a positive result in a low-probability patient has poor positive predictive value and can cause harm.

Freezing the score: probability changes as new platelet counts, imaging, cultures, coagulation results, and medication histories arrive.

Treating thrombocytopenia as a bleeding syndrome: HIT is strongly thrombotic, and decisions about anticoagulation are central.

A practical reading sequence#

Build a dated timeline of every heparin source, platelet count, surgery, transfusion, infection, thrombosis, and new medicine. Calculate percentage fall and nadir. Identify the exact onset, not the day the laboratory alerted you.

Verify thrombosis objectively where possible and list alternative causes with evidence. Score each domain separately, record uncertainty, and calculate a range when data are missing. Then follow your validated local pathway for testing and treatment.

Reassess when new information arrives. A laboratory result should update probability, not erase the clinical timeline.