DESTINY II

DESTINY II: Hemicraniectomy in Older Patients with Extensive Middle-Cerebral-Artery Stroke

DESTINY II Investigators. NEJM 2014;370:1091-1100. doi:10.1056/NEJMoa1311367

Clinical Question

  • In older patients with malignant middle cerebral artery infarction does early hemicraniectomy, compared with standard care, improve survival without severe disability?

Design

  • Randomised controlled trial
  • Randomised via online tool
  • Non-blinded
  • Intention to treat analysis for primary outcome with per-protocol sensitivity analysis
  • Sample size calculation
    • Assuming success rate of 8.6% in control group and 31% in hemicraniectomy group, 160 patients would give the study 90% power with a 5% false positive rate
  • Designed to stop recruitment as soon as harm, futility or efficacy was shown

Setting

  • 13 hospitals in Germany
  • August 2009 – May 2013

Population

  • Inclusion:
    • Age ≥61
    • Acute unilateral middle-cerebral-artery (MCA) infarction with onset of symptoms within 48 hours
    • NIHSS score >14 in patients with infarction in non-dominant hemisphere, and >19 in patients with infarction in dominant hemisphere
    • Reduced level of consciousness on NIHSS scale
    • Ischaemic infarction of at least two thirds of MCA territory, including basal ganglia on brain imaging
  • Exclusion:
    • Preexisting score of > 1 on the Modified Rankin Scale
    • Preexisting score of < 95 on the Barthel index of functional levels in activities of daily living
    • Absence of pupillary reflexes
    • GCS < 6
    • Haemorrhages or other associated brain lesions
    • Contraindications to surgery
    • Estimated life expectancy of less than 3 years
  • 112 patients enrolled
    • Comparing intervention vs. control group, baseline characteristics
      • Age 70 vs. 70
      • Pre-existing modified Rankin Score 0: 80% vs. 84%
      • GCS 12 vs. 10
      • NIHSS score 20 vs. 21

Intervention

  • Decompressive hemicraniectomy within 48 hours of onset of symptoms
    • Diameter of ≥12cm and duroplasty

Control

  • Conservative treatment
    • Options included: admission to ICU, sedation, intubation + mechanical ventilation, osmotherapy
1 patient from each group crossed over to other treatment option

Outcome

  • Primary outcome: score of 0-4 on Modified Rankin Scale at 6 months
    • 20/49 in hemicraniectomy group vs. 10/63 in control group
    • Bias-corrected, adjusted for the sequential nature of the trial
      • 38% vs. 18%, OR 2.91 (95% C.I. 1.06-7.49, P=0.04)
  • Secondary outcomes at 12 months
    • Comparing hemicraniectomy vs. control
      • Survival 57% vs. 24%
      • 6% vs. 5% had mRs score of 3 at 12 months
    • No patient who survived had a score of 0-2 on modified Rankin scale
 

Authors’ Conclusions

  • Early hemicraniectomy significantly increased probability of survival in patients >60 years of age with malignant MCA infarction, but most survivors had substantial disability

Strengths

  • Randomised controlled trial
  • Multi-centre
  • Clinical question specifically investigates older patients to compliment existing data on younger patients.

Weaknesses

  • Non-blinded – not possible to blind treating physicians or patients. Assessors of primary outcome were not part of treatment team but not told of attempts to blind them to initial treatment e.g. asking patients/relatives to not tell them what treatment they had, wearing a hat during assessment
  • Trial stopped early ‘for reasons of efficacy,’ Therefore small numbers of patients included in trial
  • It is not reported how many of the control group were admitted to ICU
  • Not powered to detect improvements resulting in mild-moderate disability only

The Bottom Line

  • In older patients with malignant MCA infarction an early decompressive hemicraniectomy significantly improves mortality but leaves the vast majority of survivors with moderately severe or severe disability. This therefore leaves us with the difficult question of whether this treatment should be offered or performed.

External Links

Metadata

Summary author: @davidslessor
Summary date: 28th May 2014
Peer-review editor: @DuncanChambler

Leave a Reply