PRECISE-DAPT Score

Clinical background

Dual antiplatelet therapy (DAPT) with aspirin and a P2Y12 inhibitor is standard after coronary stent implantation, but the duration of treatment must be weighed against bleeding risk. A longer duration, typically 12 to 24 months, reduces ischaemic events but increases bleeding complications. Short regimens of 3 to 6 months reduce bleeding but may carry ischaemic risk in high-risk patients. The decision therefore requires a structured assessment of the patient's bleeding risk before the duration of treatment is chosen. The PRECISE-DAPT score was developed precisely to meet this need: a standardised tool for estimating out-of-hospital bleeding risk and thereby guiding the choice between short and long DAPT.

Calculating the PRECISE-DAPT score

The PRECISE-DAPT score rests on five variables, each assigned points via a published nomogram:

$$\text{PRECISE-DAPT} = P_{\text{age}} + P_{\text{CrCl}} + P_{\text{Hb}} + P_{\text{WBC}} + P_{\text{bleeding}}$$

where:

• $P_{\text{age}}$: 0 to 19 points, linearly interpolated from 50 to 90 years of age (older age scores more)
• $P_{\text{CrCl}}$: 0 to 25 points, linearly interpolated from a creatinine clearance of 100 mL/min down to 0 mL/min (lower clearance scores more)
• $P_{\text{Hb}}$: 0 to 15 points, linearly interpolated from a haemoglobin of 12 g/dL down to 10 g/dL (lower haemoglobin scores more)
• $P_{\text{WBC}}$: 0 to 15 points, linearly interpolated from a white cell count of 5 to 20 ×10⁹/L (a higher count scores more)
• $P_{\text{bleeding}}$: 26 points for previous spontaneous bleeding, 0 points if none

The total score ranges from 0 to 100. The threshold for high bleeding risk is set at 25 points.

The derivation cohort consisted of 14,963 patients treated with DAPT after coronary stenting, pooled at the individual level from eight multicentre randomised trials with independent event adjudication [1]. DAPT consisted mainly of aspirin and clopidogrel, and patients with an indication for oral anticoagulation were excluded. Cox proportional hazards regression identified predictors of out-of-hospital bleeding defined as TIMI (Thrombolysis in Myocardial Infarction) major or minor. The score was validated externally in the PLATO trial (n = 8,595) and in the Bern PCI registry (n = 6,172).

Interpretation in practice

The central insight is that the score does not merely rank bleeding risk, but identifies patients in whom a long DAPT duration carries net harm. In the pooled analysis of patients randomised to different DAPT durations (n = 10,081), longer treatment (12 to 24 months) significantly increased bleeding in patients with a score ≥25, but not in those with a lower score, and the ischaemic benefit of extended treatment was seen only in the low-score group (p for interaction = 0.007) [1].

Validation and performance

In the derivation cohort a c-index for out-of-hospital bleeding (TIMI major or minor) of 0.73 (95% CI 0.61 to 0.85) was achieved [1]. In external validation in the PLATO trial the c-index was 0.70 (0.65 to 0.74) and in the Bern PCI registry 0.66 (0.61 to 0.71) [1]. Discrimination is therefore moderate and falls somewhat in real-world populations compared with the RCT cohort.

A systematic review and meta-analysis from 2023 summarised 21 studies with a total of 67,283 patients [2]. The proportion of patients at high bleeding risk (score ≥25) was 24.7%. Compared with patients with a low score, high-score patients had an odds ratio of 2.71 (95% CI 2.24 to 3.29) for out-of-hospital bleeding of any severity and 3.51 (2.71 to 4.55) for major bleeding. The pooled c-statistic for major bleeding at 1 year was 0.71 (0.64 to 0.77), confirming moderate but consistent discrimination across different populations [2].

In a comparative external validation in 4,424 ACS patients treated with prasugrel or ticagrelor, PRECISE-DAPT performed better than the PARIS bleeding score for predicting major bleeding (BARC 3 or 5), with a c-statistic of 0.653 versus 0.593 (p = 0.01) [3]. This is important because the derivation cohort was dominated by clopidogrel-treated patients, and the results suggest that the score retains some validity with more potent P2Y12 inhibitors, even though discrimination falls.

The ARC-HBR criteria (Academic Research Consortium for High Bleeding Risk), a consensus-based definition rather than a scoring model, have been compared with PRECISE-DAPT in the Bern PCI registry (12,121 patients) [4]. ARC-HBR had higher sensitivity for BARC 3 or 5 bleeding (63.8% versus 53.1% for PRECISE-DAPT) but lower specificity (62.7% versus 71.3%). PRECISE-DAPT therefore identifies fewer bleeders but with fewer false alarms, while ARC-HBR captures more at the price of a larger proportion of patients wrongly classified as high risk.

Limitations

The derivation population is dominated by clopidogrel-treated patients from RCTs, which limits generalisability to patients treated with prasugrel or ticagrelor, in whom the bleeding risk is inherently higher. The comparative validation in patients on more potent P2Y12 inhibitors showed a lower c-statistic (0.653) than in the derivation [3].

Discrimination is moderate, with a c-index around 0.70 in most validation cohorts. This means that the score alone cannot determine the treatment choice for an individual patient, but must be integrated into an overall clinical assessment.

The score does not apply to patients with an indication for oral anticoagulation, since these were excluded from the derivation cohort. In such patients the treatment situation is different, with DAPT combined with anticoagulation (triple or dual therapy), and bleeding risk is governed by other factors.

The nomogram rests on continuous variables with anchor points, and the score is sensitive to how laboratory values and creatinine clearance are calculated. At extreme values, particularly a very low creatinine clearance, the score becomes high whatever the other variables, which can lead to a patient with genuinely high ischaemic risk being recommended short DAPT on insufficient grounds.

Previous spontaneous bleeding is assigned 26 points, which on its own takes the score above the high-risk threshold. This is a strength in the sense that a history of bleeding is a strong predictor, but it also means that a patient with an otherwise low risk profile is classified as high risk on the basis of a single previous bleeding episode, whose relevance may vary.

The ARC-HBR criteria have partly replaced PRECISE-DAPT in newer guidelines as a more comprehensive tool for bleeding risk assessment, with higher sensitivity but lower specificity [4]. PRECISE-DAPT nonetheless retains the advantage of being a numerical score that is easier to use alongside ischaemic risk scores in a structured decision aid.

References

1. Costa F, van Klaveren D, James S, et al. Derivation and validation of the predicting bleeding complications in patients undergoing stent implantation and subsequent dual antiplatelet therapy (PRECISE-DAPT) score: a pooled analysis of individual-patient datasets from clinical trials. Lancet 2017;389(10073):1025-1034. PMID: 28290994
2. Munafò AR, Montalto C, Franzino M, et al. External validity of the PRECISE-DAPT score in patients undergoing PCI: a systematic review and meta-analysis. Eur Heart J Cardiovasc Pharmacother 2023;9(8):709-721. PMID: 37634083
3. Bianco M, D'Ascenzo F, Raposeiras Roubin S, et al. Comparative external validation of the PRECISE-DAPT and PARIS risk scores in 4424 acute coronary syndrome patients treated with prasugrel or ticagrelor. Int J Cardiol 2020;301:200-206. PMID: 31785951
4. Ueki Y, Bär S, Losdat S, et al. Validation of the Academic Research Consortium for High Bleeding Risk (ARC-HBR) criteria in patients undergoing percutaneous coronary intervention and comparison with contemporary bleeding risk scores. EuroIntervention 2020;16(5):371-379. PMID: 32065586