Acute intracerebral haemorrhage and diffusion-weighted imaging lesions: A meta-analysis

Diffusion-weighted imaging lesions in intracerebral haemorrhage are related to a higher risk of recurrent intracerebral haemorrhage, cognitive damage, and mortality. However, it has been reported that the relationship between the risk of diffusion-weighted imaging lesions and intracerebral haemorrhage subtype or the risk factors for diffusion-weighted imaging lesions is variable. This meta-analysis was performed to evaluate this relationship.

A systematic literature search up-to August 2020 was performed and 12 studies included 2815 subjects at the baseline with intracerebral haemorrhage. They were reporting relationships between the diffusion-weighted imaging lesions and intracerebral haemorrhage subtype or investigated the risk factors for diffusion-weighted imaging lesions. Odds ratio (OR) with 95% confidence intervals (CIs) was calculated to evaluate the prognostic role of diffusion-weighted imaging lesions and intracerebral haemorrhage subtype and investigated the risk factors for diffusion-weighted imaging lesions using the dichotomous and continuous method with a random or fixed-effect model.

Lobar intracerebral haemorrhage was not significantly related to a higher rate of diffusion-weighted imaging lesions (OR, 1.01; 95% CI, 0.75-1.36, P = .94) compared to the non-lobar intracerebral haemorrhage. Also, history of diabetes mellitus (OR, 1.15; 95% CI, 0.83-1.60, P = .39); history of smoking (OR, 0.95; 95% CI, 0.68-1.33, P = .76); history of hypercholesterolaemia (OR, 1.04; 95% CI, 0.73-1.48, P = .83) and history of ischaemic stroke (OR, 1.63; 95% CI, 0.57-4.66, P = .36) were not significantly related to higher rate of diffusion-weighted imaging lesions compared to no history of those factors. However, the history of hypertension was significantly related to a higher rate of diffusion-weighted imaging lesions (OR, 1.33; 95% CI, 1.04-1.70, P = .02) compared to no history of hypertension. Also, Subjects with diffusion-weighted imaging lesions had a greater decrease in systolic pressure in the acute phase of the intracerebral haemorrhage (OR, 10.23; 95% CI, 7.41-13.06, P < .001) compared to without diffusion-weighted imaging lesions.

Based on this meta-analysis, the history of hypertension may have an independent risk relationship with a higher rate of diffusion-weighted imaging lesions. Also, subjects with diffusion-weighted imaging lesions had a greater decrease in systolic pressure in the acute phase of the intracerebral haemorrhage compared to those without diffusion-weighted imaging lesions. This relationship forces us to recommend that identification of diffusion-weighted imaging lesions might add appreciated evidence to evaluate the progression of the underlying micro-angiopathy especially in subjects with a history of hypertension. Though further studies are needed to define the mechanisms by which these lesions may lead to cognitive damage and stroke reappearance.