Comparison of Antigen and Activation Assays with a Probability Scoring Model for the Diagnosis of Heparin Induced Thrombocytopenia.
Bibliographic record
Abstract
Background: Heparin induced thrombocytopenia (HIT) is a syndrome characterized by thrombocytopenia and an elevated risk of thrombosis. In the evaluation of patients with suspected HIT, laboratory testing is used to help differentiate between those patients who have HIT and those who do not and therefore do not require non-heparin anticoagulation. Laboratory tests include activation (serotonin release, heparin-induced platelet aggregation, ATP luminescence) and antigen (enzyme-linked immunoabsorbant assay (ELISA)) assays directed against the platelet factor 4-heparin or PF4/PVS complex. A pre-test scoring system (Warkentin 4T’s) has been derived to assess the probability of HIT which can be used in conjunction with laboratory testing in the evaluation for HIT. Objectives: To compare the activation assay used at The Ottawa Hospital (ATP luminescence) with the PF4 enhanced® ELISA and correlate results with clinical data. Methods: Patients undergoing HIT testing were identified. ELISA and activation assays were performed. Charts were reviewed in order to derive a 4T’s score and to assess response to therapeutic non-heparin anticoagulation (when administered). Correlation between ELISA, activation assay, 4T’s score and platelet response to alternative anticoagulation was determined. Sensitivity, specificity, PPV and NPV of laboratory tests or 4T’s scores were calculated depending on the measure chosen as the reference standard to diagnose HIT. Results: 111 patients undergoing HIT testing were evaluated. 41 and 43 were positive by ELISA (OD > 0.4) or activation assay respectively. 12 were positive only by ELISA (mean optical density (OD): 1.27) and 10 were positive by both activation assay and ELISA (mean OD: 2.28). Clinical information was available for 70 patients. 26 of these received therapeutic non-heparin anticoagulation as treatment for suspected HIT. Reference standard= activation assay: ELISA: Sens 95% (95% CI 73–99%), Spec 80% (66–89%), PPV 66% (46–81%), NPV 98% (86–100%). 4T’s (low vs. high score): Sens 85% (54–97%), Spec 95% (83–99%), PPV 85% (54–97%), NPV 95% (83–99%). ELISA OD > 1.5 + high 4T’s score: Sens 61% (36–82%), Spec 100% (56–100%), PPV 100% (68–100%), NPV 50% (24–76%). Reference standard= 4T’s score (high and low score only, excludes intermediate category): ELISA: Sens 85% (53–97%), Spec 77% (61–88%), PPV 52% (30–74%), NPV 94% (79–99%). Activation: Sens 85% (54–97%), Spec 95% (83–99%), PPV 85% (54–97%), NPV 95% (83–99%). Reference standard= “clear response to non-heparin anticoagulation” defined as significant platelet increase within 48 hours of start of therapy and no other explanation for platelet recovery: ELISA: Sens 100% (72–100%), Spec 54% (26–80%), PPV 68% (43–86%), NPV 100% (56–100%). Activation: Sens 85% (54–97%), Spec 69% (39–90%), PPV 73% (45–91%), NPV 82% (48–97%). 4T’s (low vs. high score): Sens 90% (54–99%), Spec 100% (60–100%), PPV 100% (63–100%), NPV 89% (51–99%). ELISA OD > 1.5 + 4T’s (low vs. high score): Sens 89% (51–99%), Spec 100% (56–100%), PPV 100% (60–100%), NPV 88% (47–99%). Conclusions: A high 4T’s score best predicts a clinical response to non-heparin therapeutic anticoagulation when HIT is suspected. ELISA OD > 1.5 does not add additional information in this respect. A negative ELISA and a low 4T’s score have comparable NPVs when an activation assay is the reference standard. Future trials should employ a clinical reference standard such as “clear response to non-heparin anticoagulation” when evaluating the operator characteristics of HIT assays.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.019 | 0.037 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".