Document resource
Primary angiitis of the central nervous system (PACNS) is a rare inflammatory disease of the cerebral or spinal vessels. It was initially described pathologically at autopsy and subsequently diagnosed by brain biopsy.1 The clinical presentation is varied and not distinctive.2–4 While histological confirmation is necessary for a definitive diagnosis, several different sets of diagnostic criteria for probable or possible PACNS have been proposed when histological confirmation is absent.4–7 These are based on clinical presentation and diagnostic tests, particularly abnormalities of the cerebral arteries visualised by angiography. In patients with clinically suspected PACNS, including those with typical angiographic abnormalities, brain biopsy shows PACNS in only 11%–33%.8–14 This is the diagnostic yield of biopsy to show PACNS in these patients, not the sensitivity. The true sensitivity is not known since that would require all patients with a negative initial biopsy to rapidly come to autopsy for the gold standard diagnostic evaluation. Importantly, brain biopsy shows an alternative diagnosis in 30%–50% including a wide variety of degenerative, demyelinating, infectious, inflammatory, neoplastic and vascular diseases. Overall, the combined diagnostic yield is 40%–83%.8–13 15 Given these data, the importance of biopsy in patients with clinically suspected PACNS cannot be overemphasised. The patients in whom the biopsy is non-diagnostic represent a dilemma. False negative brain biopsies for PACNS can occur. Specifically, case reports have described patients in whom an initial biopsy does not show PACNS when a second biopsy or autopsy does. Such cases are rare. Only 17 published cases were reported in a 2015 review, representing approximately 7% of all histologically confirmed cases reported by that time.1 16 Thus, the group with clinically suspected PACNS and a negative non-diagnostic biopsy may include an unknown proportion with PACNS mixed in with other conditions. Despite its proven diagnostic value in identifying other conditions, there remains a group of patients with clinically suspected PACNS who do not undergo biopsy. Other conditions will be present in up to one-half of these patients. This makes them very different from the group with negative biopsies from which many patients with other conditions have been excluded. Thus, patients with clinically suspected PACNS can be separated into three groups with the lowest (allowing for occasional false positives),9 lower or highest probability of other conditions based on a positive, negative or no biopsy.