Article Text

Download PDFPDF
Original research
Predictors of outcome in 1-month survivors of large middle cerebral artery infarcts treated by decompressive hemicraniectomy
  1. Barbara Casolla1,
  2. Maeva Kyheng2,
  3. Gregory Kuchcinski3,
  4. Jean-Paul Lejeune4,
  5. Riyad Hanafi3,
  6. Marie Bodenant5,
  7. Didier Leys1,
  8. Julien Labreuche2,
  9. Etienne Allart6,
  10. Merce Jourdain7,
  11. Charlotte Cordonnier1,
  12. Hilde Henon1
  1. 1 Neurology, Stroke Unit, CHU Lille, Inserm U1171, Lille, France
  2. 2 Statistics, CHU Lille, Lille, France
  3. 3 Neuroradiologie, CHU Lille, Lille, France
  4. 4 Neurosurgery, CHU Lille, Lille, France
  5. 5 Neurology, CHU Lille, Lille, France
  6. 6 Rehabilitation centre, CHU Lille, Lille, France
  7. 7 Intensive Care Unit, CHU Lille, Lille, France
  1. Correspondence to Dr Didier Leys, Neurology. Stroke unit, Univ-Lille, Inserm U1171, Lille F-59000, France; didier.leys{at}


Background Decompressive hemicraniectomy (DH) increases survival without severe dependency in patients with large middle cerebral artery (LMCA) infarcts. The objective was to identify predictors of 1-year outcome after DH for LMCA infarct.

Methods We conducted this study in consecutive patients who underwent DH for LMCA infarcts, in a tertiary stroke centre. Using multivariable logistic regression analyses, we evaluated predictors of (1) 30-day mortality and (2) poor outcome after 1 year (defined as a modified Rankin Scale score of 4–6) in 30-day survivors.

Results Of 212 patients (133 men, 63%; median age 51 years), 35 (16.5%) died within 30 days. Independent predictors of mortality were infarct volume before DH (OR 1.10 per 10 mL increase, 95% CI 1.04 to 1.16), delay between symptom onset and DH (OR 0.41, 95% CI 0.23 to 0.73 per 12 hours increase) and midline shift after DH (OR 2.59, 95% CI 1.09 to 6.14). The optimal infarct volume cut-off to predict death was 210 mL or more. Among the 177 survivors, 77 (43.5%) had a poor outcome at 1 year. Independent predictors of poor outcome were age (OR 1.08 per 1 year increase, 95% CI 1.03 to 1.12) and weekly alcohol consumption of 300 g or more (OR 5.30, 95% CI 2.20 to 12.76), but not infarct volume.

Conclusion In patients with LMCA infarcts treated by DH, stroke characteristics (infarct volume before DH, midline shift after DH and early DH) predict 30-day mortality, while patients’ characteristics (age and excessive alcohol intake) predict 1-year outcome survivors.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Twitter @leysdidier1

  • Contributors BC designed and conceptualised the study, analysed and interpreted all data and drafted the manuscript. MK and JL performed the statistical analyses. MB and GK contributed to data collection, analysed all data and revised the manuscript. MJ and EA contributed to the data collection and revised the manuscript. CC, J-PL and DL conceptualised and designed the study, interpreted study data and revised the manuscript. HH designed and conceptualised the study, analysed and interpreted all data, and revised the manuscript.

  • Funding This study was funded by Inserm U1171 and Adrinord.

  • Competing interests DL reports participation in symposia organised by Boehringer Ingelheim, Bayer, BMS and Pfizer (honoraria paid to Adrinord), drug trials with Boehringer-Ingelheim (honoraria paid to the hospital); he is vice editor of the European Stroke Journal (honoraria paid to Adrinord), vice president of the scientific committee of the Fondation de Recherche sur les AVC (unpaid); is a member of the scientific committee of the Servier Institute (unpaid); and was member of DSMBs for institutional trials (unpaid) (INCH, Germany; TARDIS, United Kingdom; and TO-ACT, the Netherlands). He received research support from a grant of the University of Heidelberg (Germany) for the ECASS4 trial. CC served for advisory boards (Bayer, Medtronics and Daiichi-Sankyo) and as investigator for clinical trials (Astra-Zeneca, Boehringer-Ingelheim, Daiichi-Sankyo and Pfizer). All fees were paid to ADRINORD or the Lille University Hospital research account; no personal funding.

  • Patient consent for publication Not required.

  • Ethics approval The study protocol was regarded as observational by the internal review board of the Lille University Hospital, which granted ethic approval for this study. Patients, or their relatives or primary caregiver, gave informed consent for the follow-up. The database was declared to the ad hoc commission protecting personal data.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data may be obtained from a third party and are not publicly available. Data may be obtained from a third party upon reasonable request and are not publicly available.

Linked Articles

  • Editorial commentary
    Karen L Furie Kevin N Sheth