Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.
Image of slice through human brain showing location of a stroke.

A new study has demonstrated that the accuracy of reporting and diagnosis of stroke types in Chinese adults is high, paving the way for future stroke research in Chinese populations.

Stroke is the leading cause of death and disability in China and accounts for two million deaths each year. Brain imaging is widely used in China to diagnose and distinguish types of stroke, but the accuracy of reporting and diagnosis had been unclear. So, researchers from the China Kadoorie Biobank Collaborative Group conducted a clinical adjudication study to assess the reporting and diagnostic accuracy of major stroke types. 

In a new report published in The Lancet Regional Western PacificDr Iain Turnbull, Clinical Research Fellow at Oxford Population Health and lead author, examined the accuracy of stroke types in the China Kadoorie Biobank. During 11 years of follow-up, over 45,000 patients were hospitalised with a stroke; medical notes were successfully retrieved and photographed in 85% of this group. This information was digitalised, reviewed and classified into stroke pathological types using World Health Organization diagnostic criteria.

Ischaemic stroke, when the blood supply to the brain is obstructed, accounted for 85% of reported strokes in this study. Haemorrhagic stroke was attributed to either intracerebral haemorrhage in 13% of strokes (when bleeding occurs inside the brain) or subarachnoid haemorrhage in 2% of strokes (when bleeding occurs on the surface of the brain). However, widespread use of brain imaging also increased the detection of 'silent lacunar infarcts' - small infarcts in the brain in the absence of typical symptoms and signs of stroke.

Overall, the diagnostic accuracy of all strokes was 81%, but was higher for strokes caused by haemorrhage (98%) than those caused by ischaemia (79%). The diagnostic accuracy of ischaemic strokes declined from 86% in 2004 - 2008 to 75% in 2016 - 2017. Importantly, the proportion of strokes due to silent lacunar infarct, increased from 7% to 18% over this period. When these silent infarcts were classified as ischaemic strokes, as advocated by recent international criteria for stroke diagnosis, the diagnostic accuracy of ischaemic strokes increased to 93%. 

Dr Turnbull said: 'This study demonstrated the feasibility of retrieval of medical records and that their clinical assessment by specialist doctors can be achieved in large-scale studies in China.’

Dr Yiping Chen, the senior author of the study who leads the adjudication programme in the China Kadoorie Biobank, said: 'The study demonstrated high levels of diagnostic accuracy of stroke types in China that were comparable with those reported in Western populations.'

The improved classification of stroke types and their subtypes should facilitate on-going and future research into prevention, prediction and treatment of stroke.