TIMI Score for unstable angina/NSTEMI

Clinical background

Patients presenting with unstable angina or NSTEMI form a heterogeneous group with a wide spread of risk for death and ischaemic events. The decision that must be made early is whether the patient should be managed invasively, with early coronary angiography and possible revascularisation, or conservatively with pharmacological treatment and a watchful approach. Without a structured risk estimate, the decision easily falls to subjective clinical judgement, risking both undertreatment of high-risk patients and unnecessary invasive investigation of patients at low risk.

The TIMI risk score for unstable angina/NSTEMI was developed to provide a simple instrument that can be calculated at the bedside and requires no computer and no laboratory test beyond a cardiac marker. The score is intended to serve two purposes: to stratify prognosis and to guide the choice between an invasive and a conservative strategy.

Calculating the TIMI risk score

The score is the sum of seven binary variables, each scoring 0 or 1:

$$\text{TIMI risk score} = x_1 + x_2 + x_3 + x_4 + x_5 + x_6 + x_7$$

where $x_i \in {0, 1}$ for:

1. Age ≥65 years
2. ≥3 risk factors for coronary artery disease (hypertension, hypercholesterolaemia, diabetes, smoking, family history)
3. Known coronary artery disease (stenosis ≥50%)
4. Aspirin in the last 7 days
5. Severe angina (≥2 episodes in 24 hours)
6. ST deviation ≥0.5 mm
7. Raised cardiac marker

The maximum score is 7. The variables were selected by multivariable logistic regression in a test cohort of 1,957 patients with unstable angina/NSTEMI from the TIMI 11B trial (August 1996 to March 1998) [1]. All the variables were independent prognostic markers for the primary endpoint: death from any cause, new or recurrent myocardial infarction, or severe recurrent ischaemia requiring urgent revascularisation within 14 days of randomisation. The score was validated in three cohorts: the unfractionated heparin arm of the ESSENCE trial and the enoxaparin arms of both TIMI 11B and ESSENCE, in total over 5,000 patients [1].

Interpretation in practice

In the derivation cohort the event rate rose progressively and significantly with increasing score, from 4.7% at a score of 0/1 to 40.9% at a score of 6/7 [1]. The clinical interpretation rests on this dose–response relationship:

The threshold at a score of 3 has acquired particular clinical importance because the derivation study demonstrated a significant interaction between the risk score and the treatment response: patients with higher scores derived greater relative benefit from more intensive treatment [1]. In clinical practice, a score ≥3 is often used as an indication for considering an early invasive strategy, while a score of 0–1 supports a more conservative approach. A score of 2 is a grey zone in which individual clinical judgement is required.

Validation and performance

In a prospective Spanish cohort of 317 patients presenting acutely with chest pain and suspected NSTE-ACS, the AUC for the TIMI risk score was 0.717 (95% CI 0.64–0.79) for predicting MACE at six weeks [2]. In the same study the HEART score performed somewhat better (AUC 0.743) and the GRACE score less well (AUC 0.649). The HEART score identified low-risk patients with higher sensitivity and a higher negative predictive value than TIMI [2].

A systematic review and meta-analysis of validation studies of risk scores in acute coronary syndromes found that the GRACE score generally performed better than TIMI over both the short and the long term [3]. The same review noted that TIMI is one of the most extensively evaluated risk scores, but that its discrimination in NSTEMI/unstable angina is lower than in STEMI, where the TIMI STEMI score shows better performance [3].

In a prospective validation in an unselected emergency department population with chest pain (1,458 patients), the TIMI risk score correlated with outcomes at 30 days, but its ability to stratify patients into clearly separated risk groups was limited [4]. Patients with a score of 0 had an event rate of 1.7% at 30 days, higher than in the derivation cohort and indicating that the score alone cannot be used to exclude ischaemic risk reliably in a broad emergency department population [4].

Limitations

The TIMI risk score was derived for and applies to patients with confirmed or strongly suspected unstable angina or NSTEMI. It is not validated for unselected chest pain patients in the emergency department, where the differential diagnostic range is broader and the pre-test probability of coronary artery disease lower [4]. Applying the score to all chest pain risks exaggerating the reassurance offered by low values.

The score does not include renal function, haemodynamic variables (blood pressure, heart rate) or heart failure class, all of which are known prognostic markers in NSTE-ACS. This is one reason why the GRACE score, which incorporates several of these variables, consistently performs better in external validations [2, 3]. In patients with a complex clinical picture, impaired renal function or haemodynamic compromise, the GRACE score should be preferred as a complementary instrument.

The cardiac marker variable is tied to the assay in use at the time of the study. With modern high-sensitivity troponin assays, more patients may be classified as having a "raised cardiac marker" than was the case in the derivation cohort, which can systematically raise the score and potentially move patients into higher risk categories. This should be borne in mind in interpretation.

The score is static and captures only the situation at presentation. A patient with an initially low score may develop new ischaemic episodes or rising cardiac markers during the admission, which warrants reassessment.

References

1. Antman EM et al. The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284(7):835–42. PMID: 10938172
2. Arispe INSR et al. Comparison of heart, grace and TIMI scores to predict major adverse cardiac events from chest pain in a Spanish health care region. Scientific Reports 2023;13:17280. PMID: 37828141
3. D'Ascenzo F et al. TIMI, GRACE and alternative risk scores in Acute Coronary Syndromes: a meta-analysis of 40 derivation studies on 216,552 patients and of 42 validation studies on 31,625 patients. Contemporary Clinical Trials 2012;33(3):507–14. PMID: 22265976
4. Chase M et al. Prospective validation of the Thrombolysis in Myocardial Infarction Risk Score in the emergency department chest pain population. Annals of Emergency Medicine 2006;48(3):252–9. PMID: 16934646