Aortic Dissection Detection Risk Score (ADD-RS)

Clinical background

Acute aortic syndrome (AAS), including aortic dissection, intramural haematoma and penetrating aortic ulcer, comprises rare but life-threatening conditions in which mortality rises by 1–2% per hour from the onset of symptoms [4]. Management in the emergency department faces a difficult balance: AAS is uncommon enough that routine CT angiography (CTA) of every patient with chest, back or abdominal pain would mean unsustainable over-investigation with radiation risk and contrast nephropathy, yet lethal enough that a missed diagnosis has catastrophic consequences. The symptoms overlap with acute coronary syndrome and pulmonary embolism, and up to 40% of cases are initially misdiagnosed [4].

The ADD-RS (Aortic Dissection Detection Risk Score) was developed as a structured tool for pre-test probability assessment, with the aim of identifying patients in whom AAS is unlikely enough that further investigation can proceed stepwise with a D-dimer, and patients in whom suspicion is high enough to proceed directly to imaging.

Calculating the Aortic Dissection Detection Risk Score

The score rests on three clinical categories, each assessed as present (1) or absent (0):

$$\text{ADD-RS} = \mathbb{1}(\text{high-risk condition}) + \mathbb{1}(\text{high-risk pain}) + \mathbb{1}(\text{high-risk examination finding})$$

The total gives a value between 0 and 3. The categories are:

• High-risk conditions: Marfan syndrome or another connective tissue disease, a family history of aortic disease, known aortic valve disease, recent aortic manipulation, or a known thoracic aortic aneurysm.
• High-risk pain features: chest, back or abdominal pain of abrupt onset, severe intensity, or tearing or ripping in character.
• High-risk examination findings: a pulse deficit or difference in systolic blood pressure, a focal neurological deficit with pain, a new diastolic murmur over the aorta, or hypotension or shock.

The derivation cohort consisted of 2,538 patients with confirmed acute aortic dissection in the International Registry of Acute Aortic Dissection (IRAD), enrolled between 1996 and 2009 [1]. The score was constructed from the 12 clinical risk markers identified in the 2010 ACC/AHA guidelines for thoracic aortic disease and tested retrospectively for sensitivity: 95.7% of patients had at least one risk marker [1].

Interpretation in practice

ADD-RS is not used as a stand-alone diagnostic test, but as one step in a diagnostic algorithm in which the next step is governed by the score and possibly the D-dimer.

The clinically most important point is that an ADD-RS ≥2 means the D-dimer must not be used as an exclusion test. A failure rate of 4.2% for a negative D-dimer at an ADD-RS >1 has been reported from the ADvISED data, which is unacceptable [5].

Validation and performance

In the derivation cohort the patients were distributed as follows: 4.3% had an ADD-RS of 0, 36.5% an ADD-RS of 1, and 59.2% an ADD-RS of 2–3 [1]. The score was designed for sensitivity, not specificity, which is decisive for interpreting all the subsequent validation studies.

The prospective multicentre ADvISED study (1,850 patients at 6 hospitals in 4 countries, 2014–2016; AAS prevalence 13%) showed that a strategy of ADD-RS 0 plus a negative D-dimer gave a failure rate of 0.3% (95% CI 0.1–1.9) with an efficiency of 15.9% [2]. The strategy of ADD-RS ≤1 plus a negative D-dimer gave the same failure rate, 0.3% (95% CI 0.1–1.0), but with a considerably higher efficiency of 49.9% [2].

A hierarchical meta-analysis including 13 studies (of which 6 evaluated ADD-RS combined with a D-dimer) found the following pooled diagnostic performance [3]:

The last strategy gives the best balance between safety and resource use according to a decision analysis in the same project [4]. At a prevalence of 0.26% in an unselected population, however, no strategy involving a D-dimer was cost-effective; only when clinicians had selected a population with an AAS prevalence of 0.61% did the combination of ADD-RS >1 or (ADD-RS =1 and D-dimer >500) become both cost-effective and feasible [4].

A retrospective study from Auckland (181 patients with an ADD-RS ≤1 who underwent CTA, 2009–2019) found a failure rate of 0% for a negative D-dimer, but with an upper 95% CI limit of 3.3%, which exceeds the proposed acceptable threshold of 0.5% [5]. The prevalence of AAS was only 5.0% and a D-dimer had been requested in only 11%, so the confidence interval is wide and the finding should be interpreted with caution [5].

Limitations

ADD-RS was derived from a registry of confirmed cases, not from a prospective cohort of patients with suspected AAS. This means that the score measures sensitivity but lacks a true control group in the derivation. The low specificity, confirmed in the meta-analysis as 34.7% for ADD-RS >0, means that the majority of patients without AAS nonetheless score ≥1 [3].

According to several studies, the score cannot reliably distinguish low from intermediate risk, that is, ADD-RS 0 from ADD-RS 1 [5]. This is clinically relevant because both levels can, according to ADvISED, be combined with a negative D-dimer for exclusion, but with different efficiency: the strategy that includes ADD-RS 1 spares considerably more patients a CTA [2].

ADD-RS applies to patients in whom AAS is a genuine differential diagnosis. Applying the score to every patient with chest pain in an unselected population leads to a high rate of unnecessary CTA, since its specificity is too low to serve as a broad screening tool [4].

One addition that may improve performance is point-of-care ultrasound (POCUS). In the PROFUNDUS study, a prospective multicentre study of 1,979 patients at 12 emergency departments, integrating POCUS with ADD-RS gave a net reclassification of 20% compared with the clinical score alone, and no patient who met the exclusion criteria developed AAS within 30 days (95% CI 0–0.41) [6].

References

1. Rogers AM et al. Sensitivity of the aortic dissection detection risk score, a novel guideline-based tool for identification of acute aortic dissection at initial presentation: results from the international registry of acute aortic dissection. Circulation 2011;123(20):2213–8. PMID: 21555704
2. Nazerian P et al. Diagnostic Accuracy of the Aortic Dissection Detection Risk Score Plus D-Dimer for Acute Aortic Syndromes: The ADvISED Prospective Multicenter Study. Circulation 2018;137(3):250–8. PMID: 29030346
3. Ren S et al. Diagnostic accuracy of the aortic dissection detection risk score alone or with D-dimer for acute aortic syndromes: Systematic review and meta-analysis. PLoS One 2024;19(6):e0304401. PMID: 38905181
4. Goodacre S et al. Diagnostic strategies for suspected acute aortic syndrome: systematic review, meta-analysis, decision-analytic modelling and value of information analysis. Health Technol Assess 2025;29(45). PMID: 40944621
5. Bhat S et al. d-Dimer With the Aortic Dissection Detection Risk Score May Improve Patient Selection for Acute Aortic Syndrome Diagnostic Imaging. Emerg Med Australas 2026;38(1):e70217. PMID: 41582603
6. Morello F et al. Diagnosis of acute aortic syndromes with ultrasound and d-dimer: the PROFUNDUS study. Eur J Intern Med 2024;128:94–103. PMID: 38871565