Article Text


Quality of life after decompressive craniectomy for malignant middle cerebral artery infarction
  1. K Vahedi1,
  2. L Benoist1,
  3. A Kurtz1,
  4. J Mateo2,
  5. A Blanquet3,
  6. M Rossignol4,
  7. P Amarenco5,
  8. A Yelnik6,
  9. E Vicaut7,
  10. D Payen8,
  11. M G Bousser9
  1. 1Service de Neurologie, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  2. 2Départment d’Anesthésie-Réanimation, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  3. 3Service de Neurochirurgie, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  4. 4Départment d’Anesthésie-Réanimation, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  5. 5Service de Neurologie, Hôpital Bichat, AP-HP, Paris, France
  6. 6Service de Médecine Physique et de Réadaptation, Hôpital Lariboisière-Fernand Widal, AP-HP, 200 rue du Faubourg, Saint-Denis, 75010 Paris, France
  7. 7Unité de Recherche Clinique, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  8. 8Départment d’Anesthésie-Réanimation, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  9. 9Service de Neurologie, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France
  1. Correspondence to:
 Dr Katayoun Vahedi
 Service de Neurologie, Hôpital Lariboisière, AP-HP, 2 rue A. Paré, 75010 Paris, France;

Statistics from

Malignant middle cerebral artery (MCA) infarction is a devastating condition leading to early death in nearly 80% of cases due to the rapid rise of intracranial pressure despite maximum medical management of the ischaemic brain oedema.1 Decompressive craniectomy (DC) has been proposed to prevent brain herniation in malignant MCA infarction, but it remains controversial in the absence of randomised controlled trials and because of the fear of a severe residual disability after surgery.1–4 We present herein the results of a quality of life assessment using patient and proxy versions of the Stroke Impact Scale (SIS) in eight patients 12–30 months after craniectomy for malignant MCA infarcts.


Between March 1999 and November 2000, all consecutive patients with malignant MCA infarction were treated by DC and durotomy at Lariboisière Hospital if they were younger than 55 years of age, had a complete MCA infarct as defined by complete MCA territory CT ischaemic changes, and a severe hemiplegia with altered level of consciousness with further neurological deterioration due to brain oedema, and if a close family member gave informed consent. Exclusion criteria were: prestroke moderate to severe disability defined by a modified Rankin scale (mRS)⩾2, haemorrhagic transformation involving more than 50% of the MCA territory, and significant contralateral ischaemia.

Disability was assessed using the mRS and the Barthel Index (BI), and quality of life using the French version of the SIS 2.0.5 The SIS comprises eight domains, four physical domains (including strength, hand function, mobility, and activity of daily living/instrumental activities of daily living) and four psychosocial domains (including emotion, communication, memory, and social participation) and includes the patient’s global assessment of percentage of recovery. The scores of each domain range from 0 to 100, with 100 being the best.


Ten patients were included (eight men and two women, mean (SD) age 41 (12) years, range 15–54). The mean (SD) NIH score scale at admission was 21 (3), range 16–25. Five patients had a left sided stroke with severe aphasia. The mean time between stroke onset and surgery was 65 (68) h, range 12–252. One patient had a late DC because of recurrent MCA infarct at day 9 after the first stroke. All patients had signs of temporal herniation before surgery including uni- or bilateral mydriasis (9/10), Cheynes-Stokes hypoventilation (8/10), or decerebration (6/10). The mean (SD) duration of hospitalisation in the intensive care unit was 22 (20) days, range 3–58. Two patients died, one from a cerebral abscess and the other from a large epidural hematoma.

All living patients (8/10) were followed for a mean (SD) duration of 21 (21) months, range 12–30. All were managed in a specialised stroke rehabilitation unit with a mean (SD) hospital stay of 12 (11) months, range 4–24, after which they returned home with either home rehabilitation facility or day hospital care. At the end of follow up, 7/8 patients had an mRS⩽4 (table 1). The mean (SD) NIH score scale was 13 (4), range 8–18. The two youngest patients had the best scores on disability (mRS = 2) and were fully independent for the activities of daily living (BI ⩾90) (table 1).

Table 1

 Domain scores of the SIS questionnaire filled in by seven living patients and eight proxies 12–30 months after decompressive craniectomy

The 64 SIS items could be measured in all patients except patient 7 who had severe aphasia (table 1). The proxy version of the SIS was administered to a close relative (five spouses, two parents) or an employed caregiver (one). The mean (SD) patient assessment of global perception of stroke recovery was 59 (16). The score was lower, but not significantly so, in patients with aphasia compared to patients without, both in patient (55 (15) v 65 (19), p = 0.48, Wilcoxon test) and proxy (49 (17) v 57 (18), p = 0.45, Wilcoxon test) versions of the measurement. The combined mean (SD) physical domain recovery was 48 (16) when assessed by patients and 39 (16) when assessed by proxies. The lowest scaling success rate was for hand function and the highest for emotion domain recovery. However, during the follow up, two patients had a major depressive episode. In addition, one spouse attempted suicide (patient 8). As expected, patients with aphasia had a lower mean (SD) rate of recovery for communication (50 (37)) than those without (91 (14)), although the difference was not statistically significant (p = 0.21, Wilcoxon test). No patient returned to his or her prior employment, although one patient, the youngest (patient 3), returned to school.


This study shows that the SIS measurement is applicable to patients with malignant MCA infarction 12–30 months after craniectomy. The patient’s assessment of the physical aspects of disability at 12–30 months post stroke was high (all physical domains mean recovery of 48/100). Interestingly, the proxy assessment of physical domains recovery was lower (39/100) than the patient assessment. In addition, the disability measured by the mRS showed that 6/8 living patients had an mRS>3, which may indicate a poor outcome. It may be that in patients with malignant MCA infarction, the patient version of the SIS overestimates the physical recovery because of cognitive dysfunction including unilateral neglect, anosognosia, or aphasia.

One main concern in malignant MCA infarction is the psychosocial impact of stroke. In our study, the percentage of recovery was good for emotion and memory but moderate for communication and participation. As expected, patients with aphasia had a lower rate of recovery for communication than patients without, though the difference did not reach statistical significance, presumably because of the small numbers. In the same way, the global percentage of recovery was lower, but not significantly, in patients with aphasia than in patients without. Interestingly, the proxy’s assessment of psychosocial recovery, though lower, was close to the patient’s assessment.

In conclusion, this study shows that after craniectomy for malignant MCA infarcts, even though the perception of physical aspects of disability is high, that of psychosocial impairment is lower. Open series of craniectomy for malignant MCA infarction indicate that surgery decreases death rates. However, randomised trials are needed, taking into account not only death and dependency but also quality of life.


The authors thank Mapi Research Institute, Lyon, France for providing the French version of the Stroke Impact Scale 2.0 (Duncan et al, copyright University of Kansas Medical Center) and Olivier Chassany from the Délégation à la Recherche Clinique (AP-HP) for his helpful comments.


View Abstract


  • Competing interests: none declared

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.