Review of the ADvISED Prospective Multicenter Study
by: Sam Du, PGY-3
Background:
- Acute aortic syndromes (AAS):
- aortic dissection,
- intramural aortic hematoma,
- penetrating aortic ulcer,
- aortic rupture
- 6/100,000 individuals/year
- Misdiagnosis rate: 14-39% and high mortality indicates low sensitivity of gestalt and need for high sensitivity diagnostic algorithm
- Only 2.7% of CT aortograms are positive
- Other imaging methods: MRA and TEE (but rarely used)
- Aortic dissection detection risk score (ADD-RS):
- Retrospectively derived based on common risk factors identified in patients in the International Registry for Acute Aortic Dissection (IRAD) database
- Supported by guidelines but not prospectively validated
- Scored 0-3 based on the presence of risk factors in 3 categories with a maximum point of 1 in each category:
- Known aortic disease, recent aortic manipulation or possible familial connective tissue disease
- Severe, abrupt, or ripping/tearing pain
- Shock, neurovascular deficit, or new murmur
- D dimer alone is not sensitive enough
- Meta analysis of prospective studies using D dimer cut off of 0.5 mcg/mL showed sensitivity of 95-98%
Study design
- Multicenter, prospective, observational study from 2014-2015
- Patients were consecutively screened from 6 hospitals in 4 countries none of them in the US
- Aged > 18 years, Symptoms <=14 days, One or more of chest pain, abdominal pain, back pain, syncope, neurovascular deficit
- Symptoms <=14 days is the standard arbitrary definition of acute AAS
- “Patients were included only if AAS was considered in the differential diagnosis by the attending physician”
- Criteria was not defined
- Number of patients screened but not included is not reported
- Included patients initially risk stratified into groups by ADD-RS score of 0,1, or > 1 and then each group was further stratified by D-dimer positive of negative
- Physicians decided whether to image using their own gestalt, ADD-RS, and D-dimer
- Patients that were not imaged were followed for 14 days for AAS diagnosis, revisit, death
- A two physician panel blinded to ADD-RS and D-dimer results reviewed all cases and identified AAS present or not-present based on imaging, surgery, or autopsy
Primary results
- Of the patients included in the study 241/1850 (13%) patients were found to have AAS
- Much higher than a typical ED cohort given historically only 2.7% of CT aortograms are positive
- Using ADD-RS = 0 with a negative D-dimer:
- 294/1850 (16%) patients were ruled out without imaging, while missing 1 case
- Missed case rate of 0.3% (Sensitivity 99.6%)
- Specificity 18.2%
- 78 y/o F PMHx HTN, DM, smoker p/w posterior CP x 7 d with widened mediastinum on CXR found to have Type B dissection with an ADD-RS=0 and a negative D-Dimer would have been missed
- Using ADD-RS <=1 with a negative D-dimer:
- 924/1850 (50%) patients were ruled out without imaging while missing 3 cases
- Missed case rate of 0.3% (Sensitivity 98.8%)
- Specificity 57.3%
- 72 y/o M PMHx HTN, CAD p/w sudden severe ripping anterior CP and syncope found to have Type A dissection with an ADD-RS=1
- 34M no PMH p/w Sudden severe ripping posterior CP and syncope with widened mediastinum on CXR found to have Type A dissection with an ADD-RS=1
- Given the similar sensitivities and miss rates, but a greater specificity the authors propose ADD-RS <= 1 with a negative D-dimer as a rule out criteria for AAS
Additional results
- D-dimer alone had a sensitivity of 97% and specificity of 64%, missing 8 cases
- Median level of D-dimer in AAS was 5.810 mcg/mL
- ADD-RS alone with a cut off of <=1 had a sensitivity of 95% and specificity of 26
- Significant risk factors:
- Age,
- HTN, and
- AAA
- Significant historical factors:
- Back pain,
- abdominal pain,
- syncope, and neurovascular deficit
- Sudden or severe pain
- Any suggestive physical exam finding had an odds ratio of 7
Strengths
- Prospective multicenter
- Large study of a rare and difficult to study diseases
- Built in case control study of risk factors for AAS
Weaknesses
- No hospitals in the US limits external validity
- Patients were included based on the attending’s gestalt
- Limits external validity as their gestalt is not reproducible making daily application to clinical practice difficult
- Likely has a negligible effect on conclusions on sensitivity as it is unlikely a substantial number of missed cases were excluded from the study
- Limits conclusions on specificity as gestalt already excluded the most low risk patients
- Effects are unknown as the total number of missed cases is not reported and the number of excluded patients is not reported
- The number of excluded patients is likely very high since the rate of AAS in the study population is also very high at 13% compared to < 4% in all comers with chest pain
- Probably necessary to achieve buy in from clinicians and avoid overtesting of D-dimer
- Though still 47% of patients were imaged
- If pulmonary embolism is the true diagnosis a CT aortogram may miss more subtle emboli
- Relies on ADD-RS which was not previously prospectively validated
- The three cases highlighted above had low ADD-RS score (since multiple suggestive historical are only counted under one point in ADD-RS)
- This was an arbitrary decision made in the guidelines
- Historical features were among the most common risk factors among patients in the IRADS database
- The risk factors found in this study but not in ADD-RS may have additional clinical value
- Syncope
- HTN and age
- Pain radiating to the back or above and below the diaphragm
- Widened mediastinum is included in the AHA guideline algorithm but not included in the ADD-RS
- The three cases highlighted above had low ADD-RS score (since multiple suggestive historical are only counted under one point in ADD-RS)
- Observational study. RTC is needed to compare new algorithm with gestalt
- The three cases highlighted above were missed by the algorithm but would unlikely to be missed with gestalt
- Providers were not blinded to ADD-RS or D-Dimer
- A proxy for gestalt may be the 1013 that were sent home without imaging. Likely a lower risk group than those that were imaged. 7 AAS were found in this group. Suggesting a miss rate for gestalt 0.7% compared to 0.3% of the proposed algorithm. All missed cases were diagnosed on revisit, none were diagnosed at autopsy.
- Consider using age-adjusted D-dimer in future studies to further increase specificity
- Follow up length of 14 days may not be sufficiently long
- Definition of acute AAS as <= 14 days is arbitrary. An acute AAS that a patient survives for > 14 days is still clinically significant and should be diagnosed
- No standardized acceptable miss rate of AAS
References:
- Diagnostic Accuracy of the Aortic Dissection Detection Risk Score Plus D-Dimer for Acute Aortic Syndromes: The ADvISED Prospective Multicenter Study. Nazerian P, Mueller C, Soeiro AM, Leidel BA, Salvadeo SAT, Giachino F, Vanni S, Grimm K, Oliveira MT Jr, Pivetta E, Lupia E, Grifoni S, Morello F; ADvISED Investigators. Circulation. 2018 Jan 16;137(3):250-258. doi: 10.1161/CIRCULATIONAHA.117.029457. Epub 2017 Oct 13. PMID: 29030346
- 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. Rogers AM, Hermann LK, Booher AM, Nienaber CA, Williams DM, Kazerooni EA, Froehlich JB, O’Gara PT, Montgomery DG, Cooper JV, Harris KM, Hutchison S, Evangelista A, Isselbacher EM, Eagle KA; IRAD Investigators. Circulation. 2011 May 24;123(20):2213-8. doi: 10.1161/CIRCULATIONAHA.110.988568. Epub 2011 May 9. PMID: 21555704
- 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Hiratzka LF, et al.; American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines; American Association for Thoracic Surgery; American College of Radiology; American Stroke Association; Society of Cardiovascular Anesthesiologists; Society for Cardiovascular Angiography and Interventions; Society of Interventional Radiology; Society of Thoracic Surgeons; Society for Vascular Medicine. Circulation. 2010 Apr 6;121(13):e266-369