The Boston criteria 2.0 were proposed in 2022 in order to better include leptomeningeal and white matter characteristics into the diagnoses of probable and possible cerebral amyloid angiopathy (CAA) ¹. They consist of combined clinical, imaging and pathological parameters, and are based upon the original Boston criteria and modified Boston criteria, which were proposed in 1995 and 2010 respectively ^2,3.

It is not to be confused with the Boston criteria for blunt cerebrovascular injury.

Criteria

The criteria are divided into four tiers ¹:

• definite CAA

  • full brain post-mortem examination demonstrating:

    • spontaneous intracerebral haemorrhage, transient focal neurological episodes, convexity subarachnoid haemorrhage, or cognitive impairment or dementia

    • severe CAA with vasculopathy

    • absence of other diagnostic lesion

• probable CAA with supporting pathology

  • clinical data and pathological tissue (evacuated haematoma or cortical biopsy) demonstrating:

    • presentation with spontaneous intracerebral haemorrhage, transient focal neurological episodes, convexity subarachnoid haemorrhage, or cognitive impairment or dementia

    • some degree of CAA in specimen

    • absence of other diagnostic lesion

• probable CAA

  • pathological confirmation not required

  • for patients aged 50 years and older

  • presentation with spontaneous intracerebral haemorrhage, transient focal neurological episodes, or cognitive impairment or dementia

  • MRI criteria:

    • demonstrates either:

      • at least two of the following strictly lobar haemorrhagic lesions on T2*-weighted MRI, in any combination: intracerebral haemorrhage, cerebral microbleeds, or foci of cortical superficial siderosis (multiple distinct foci are counted as independent haemorrhagic lesions) or convexity subarachnoid haemorrhage (multiple distinct foci are counted as independent haemorrhagic lesions);
        OR

      • one lobar haemorrhagic lesion plus one white matter feature (severe perivascular spaces in the centrum semiovale [>20 visible in one hemisphere] or white matter hyperintensities in a multispot pattern [>10 subcortical FLAIR dots bilaterally])

    • in the absence of:

      • any deep haemorrhagic lesions on T2*-weighted MRI
        AND

      • other cause of haemorrhagic lesions*

    • haemorrhagic lesion in cerebellum not counted as either lobar or deep haemorrhagic lesion

• possible CAA

  • pathological confirmation not required

  • for patients aged 50 years and older

  • presentation with spontaneous intracerebral haemorrhage, transient focal neurological episodes, or cognitive impairment or dementia

  • MRI criteria:

    • demonstrates either:

      • one strictly lobar haemorrhagic lesion on T2*-weighted MRI: intracerebral haemorrhage, cerebral microbleeds, or foci of cortical superficial siderosis or convexity subarachnoid haemorrhage;
        OR

      • one white matter feature (severe perivascular spaces in the centrum semiovale [>20 visible in one hemisphere] or white matter hyperintensities in a multispot pattern [>10 subcortical FLAIR dots bilaterally])

    • in the absence of:

      • any deep haemorrhagic lesions on T2*-weighted MRI
        AND

      • other cause of haemorrhagic lesions*

    • haemorrhagic lesion in cerebellum not counted as either lobar or deep haemorrhagic lesion

* "Other cause of haemorrhagic lesions" are: antecedent head trauma, haemorrhagic transformation of an ischaemic stroke, arteriovenous malformation, haemorrhagic tumour, warfarin therapy with INR >3, and vasculitis ¹.

In its initial publication, the Boston criteria 2.0 for diagnosis of 'probable CAA' was validated against multiple different cohorts, and against patients who had had autopsy it had a sensitivity of 74.5% (95% confidence interval (CI) 65.4% to 82.4%) and specificity of 95% (95% CI 83.1% to 99.4%), which is superior compared to when the modified Boston criteria were applied to the same patient cohorts ¹. Notably, when applied to patients who are asymptomatic or only have cognitive impairment, diagnostic accuracy of the Boston criteria 2.0 is much lower, with a diagnosis of 'probable CAA' only having a sensitivity of 28.6% (95% CI 13.2% to 48.7%) and specificity of 65.3% (95% CI 44.3%–82.8%) ⁴.