Preoperative MRI characteristics predict chronic subdural haematoma postoperative recurrence: a meta-analysis

Abstract

Purpose

Prior studies have reported that preoperative T1 magnetic resonance imaging (MRI) signal iso- or hypointensity may indicate higher risk of postoperative chronic subdural haematoma recurrence (cSDH). The authors undertook a meta-analysis to determine whether preoperative MRI characteristics may predict recurrence and/or reoperation after initial surgical evacuation of cSDH.

Materials and methods

Embase, PubMed and Cochrane Library were queried to find articles published after 1990 that included data on preoperative brain MRIs obtained prior to burr hole or craniotomy haematoma evacuation of unilateral or bilateral cSDH and data on postoperative recurrence and/or repeat evacuation. The authors specifically investigated the T1 signal characteristics of the haematoma as they related to postoperative recurrence.

Results

Five articles were identified that included preoperative MRI T1 signal characteristics and postoperative recurrence data. One study reported cSDH recurrence requiring reoperation as the primary outcome, whereas four studies reported SDH recurrence alone as the primary outcome. A total of 1081 patients with a total of 1290 cSDHs underwent surgical evacuation. In the combined analysis, there were 62 recurrences in 300 cases (20.7%) in the MRI T1 hypo- and/or iso-intensity groups and 59 recurrences in 885 cases (6.7%) in the MRI T1 other groups (combined odds ratio = 4.385 (95% CI 2.93–6.57)). There was low heterogeneity among studies (i² = 0%).

Conclusion

This meta-analysis suggests that preoperative MRI T1 hypo- or isointensity cSDH signal may predict increased postoperative SDH recurrence risk.

Keywords:

• Chronic subdural haematoma
• magnetic resonance imaging
• meta-analysis
• recurrence
• surgery

Acknowledgments

We thank Kristin Kraus, MSc, for editorial assistance with manuscript preparation.

Disclosure statement

Scott McNally is supported by AHA Scientist Development Grant 17SDG33460420 (PI McNally) and NIH Grant 1R01 HL127582-01A1 (PI Parker). Ramesh Grandhi is a consultant for Medtronic Neurovascular, Cerenovus, and BALT Neurovascular. The other authors report no conflict of interest.

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.