Age Has a Crucial Effect on Outcome After Hemicraniectomy

Abstract & Commentary

Sources: Uhl E, et al. Outcome and prognostic factors of hemicraniectomy for space occupying cerebral infarction. J. Neurol Neurosurg Psychiatry. 2004;75:270-274; Gupta R, et al. Hemicraniectomy for massive middle cerebral artery territory infarction: A systematic review. Stroke. 2004;35: 539-543.

Approximately 10% of all patients with middle cerebral artery (MCA) territory cerebral infarction suffer progressive deterioration due to cerebral edema, increased intracranial pressure, and brain herniation.1 In these patients with malignant MCA territory infarction aggressive decompressive craniectomy has been reported to improve overall prognosis.2 Several case series or case reports have reported improved survival of patients after craniectomy, but small sample size, poorly defined selection criteria, and absence of control groups prevent a definitive conclusion as to any benefit in functional outcomes.

Uhl and colleagues and Gupta and colleagues each performed a review of the literature to identify patients most likely to benefit from hemicraniectomy. Uhl et al identified 188 patients treated between 1996 and 2001 whom they considered for prognostic evaluation. Gupta et al identified 129 patients in the English language literature and added 9 patients from New York-Presbyterian Hospital-Columbia University Medical Center. All studies reviewed were uncontrolled case series; there has been no randomized, controlled trial of hemicraniectomy.

Both groups used the Glasgow Outcome Scale (GOS) to describe outcome3 and, not surprisingly, reported similar results. Uhl et al found that at 3, 6, and 12 months mortality rates were 8%, 38%, and 44%, respectively. Patients younger than 50 without infarction involving a vascular territory in addition to the MCA territory had an adjusted 6-month mortality of only 20%. In contrast, patients older than 50 with involvement of one or more vascular territories had a 60% mortality at 6 months. Thirty-five percent of patients younger than 50 achieved a good recovery or had only mild or moderate disability as defined by the GOS, while only 12% of the elderly population did so.

Gupta et al found that after a minimum of 4 months, 10 patients (7%) were functionally independent (good recovery) and 48 (35%) had died or were severely disabled. Of 75 patients who were older than 50, 80% were dead or severely disabled compared with 32% of 63 patients younger than 50 (P < .00001). The timing of surgery, hemisphere infarcted, presence of signs of herniation before surgery and, in contrast to the finding of Uhl et al, involvement of other vascular territories did not affect outcome. Both groups of investigators concluded that age is the most important pretreatment prognostic factor.


Both reviews, while informative, are limited by the fact that the data analyzed were obtained from uncontrolled, retrospective studies of hemicraniectomy patients. The patients, as well as the studies, were heterogeneous and lacked a uniform surgical approach, time of follow-up, and prognosis. Nevertheless, the reports underscore the point that age must be considered the most important pretreatment prognostic factor. The results of hemicraniectomy are encouraging, although the benefit of surgery for patients older than 50 is questionable. Future trials should focus on hemicraniectomy in younger patients and on early decompression before there are clinical signs of brain herniation as a means of improving functional outcome. Two prospective trials are under way and should provide guidelines for the timing of surgery.4,5 — John J. Caronna

Dr. Caronna, Vice-Chairman, Department of Neurology, Cornell University Medical Center; Professor of Clinical Neurology, New York Hospital is Associate Editor of Neurology Alert.


1. Moulin DE, et al. Stroke. 1985;16:282-284.

2. Hacke W, et al. Arch Neurol. 1996;53:309-315.

3. Jennett B, et al. Lancet. 1975;1:480-484.

4. Frank JL, et al. Stroke. 1999;30:243. Abstract.

5. Hotmeyer J, et al. Cerebrovasc Dis. 2001; 11(suppl 4):34. Abstract.