Background: Acute aortic syndrome is a life-threatening condition that requires urgent diagnosis with computed
tomographic angiography. Diagnostic technologies, including clinical scores and biomarkers, can be used to select
patients presenting with potential symptoms of acute aortic syndrome for computed tomographic angiography.
Objectives: We aimed to estimate the accuracy of clinical scores and biomarkers for diagnosing acute aortic
syndrome, the cost-effectiveness of alternative diagnostic strategies and the expected value of future research.
Methods: We searched online databases from inception to February 2024, reference lists of included studies and
existing systematic reviews. We included cohort studies evaluating the accuracy of clinical scores or biomarkers for
diagnosing acute aortic syndrome compared with a reference standard. Two authors independently selected and
extracted data. Risk of bias was appraised using the quality assessment of diagnostic accuracy studies-2 tool. Data
were synthesised using either a multinomial or a bivariate normal meta-analysis model.
We developed a decision-analytic model to simulate the management of a hypothetical cohort of patients attending
hospital with possible acute aortic syndrome. We modelled diagnostic strategies that used the Aortic Dissection
Detection Risk Score and D-dimer to select patients for computed tomographic angiography. We used estimates
from our meta-analysis, existing literature and clinical experts to model the consequences of diagnostic strategies
upon survival, health utility and healthcare costs. We estimated the incremental cost per quality-adjusted life-year
gained by each strategy compared to the next most effective alternative on the efficiency frontier, and the expected
value of perfect information.
Results: Primary meta-analysis included 12 studies of Aortic Dissection Detection Risk Score alone, 6 studies of
Aortic Dissection Detection Risk Score with D-dimer and 18 studies of D-dimer using the 500 ng/ml threshold.
Sensitivities and specificities (95% credible intervals) were: Aortic Dissection Detection Risk Score > 0 94.6% (90%
to 97.5%) and 34.7% (20.7% to 51.2%), Aortic Dissection Detection Risk Score > 1 43.4% (31.2% to 57.1%) and
89.3% (80.4% to 94.8%); Aortic Dissection Detection Risk Score > 0 or D-dimer > 500 ng/ml 99.8% (98.7% to 100%)
and 21.8% (12.1% to 32.6%); Aortic Dissection Detection Risk Score > 1 or D-dimer > 500 ng/ml 98.3% (94.9% to
99.5%) and 51.4% (38.7% to 64.1%); Aortic Dissection Detection Risk Score > 1 or Aortic Dissection Detection Risk
Score = 1 with D-dimer > 500 ng/ml 93.1% (87.1% to 96.3%) and 67.1% (54.4% to 77.7%); and D-dimer alone 96.5%
(94.8% to 98%) and 56.2% (48.3% to 63.9%). We identified 11 cohort studies of other biomarkers, but accuracy
estimates were limited and inconsistent.
Decision-analytic modelling showed that applying diagnostic strategies to an unselected population (acute aortic
syndrome prevalence 0.26%) resulted in high rates of computed tomographic angiography, and only the strategy
selecting patients with Aortic Dissection Detection Risk Score > 1 for computed tomographic angiography was costeffective. If clinicians can select a population for investigation with higher acute aortic syndrome prevalence (0.61%),
then using a strategy of Aortic Dissection Detection Risk Score > 1 or Aortic Dissection Detection Risk Score = 1
with D-dimer > 500 ng/ml or a strategy of Aortic Dissection Detection Risk Score > 1 or D-dimer > 500 ng/ml to
select patients for computed tomographic angiography is cost-effective and deliverable. At a threshold of £20,000/
quality-adjusted life-year, population expected value of perfect information was around £17.75M.
Limitations: Studies included in the meta-analysis showed substantial heterogeneity in estimates of specificity. In
the modelling, there was substantial uncertainty around what constitutes suspected acute aortic syndrome and the
effect of delayed diagnosis.
Conclusions: The Aortic Dissection Detection Risk Score and D-dimer provide useful diagnostic information and
may offer cost-effective strategies for selecting patients for computed tomographic angiography, but their role
depends upon how clinicians identify suspected acute aortic syndrome.
Future work: Primary research is required to compare different combinations of Aortic Dissection Detection
Risk Score with D-dimer in practice, explore how suspected acute aortic syndrome is identified and evaluate
alternative biomarkers.
Funding: This synopsis presents independent research funded by the National Institute for Health and Care Research
(NIHR) Health Technology Assessment programme as award number NIHR151853.
A plain language summary of this synopsis is available on the NIHR Journals Library Website https://doi.org/10.3310/
GGOP6363.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)