In the early 2000s, three trials were independently conducted in Europe which compared decompressive hemicraniectomy with standard medical therapy within 48 h of presentation for patients under 60 years of age who developed clinical deterioration following middle cerebral artery infarction. The DECIMAL and DESTINY trials were interrupted early because of slow recruitment and a significant difference in mortality between the treatment groups favouring surgery [2, 3]. The HAMLET trial was stopped because it was thought that it would be highly unlikely that a statistically significant difference would be seen for the primary neurological outcome measure which was defined as good (modified Rankin Score [mRS] 0–3) or poor (mRS 4–6) [4].

What each trial independently demonstrated was that survival was improved in those patients who were randomised to receive decompressive surgery. However each individual trial was insufficiently powered to determine whether there was an improvement in favourable outcome. Thereafter, a pooled analysis of the 93 patients involved in all three trials was performed, and this confirmed not only the significant reduction in mortality, but it was also concluded that there was an increase in the number of patients with a favourable outcome (Table 13.1).

On initial examination, this would appear to provide compelling evidence for the efficacy of surgical decompression for patients less than 60 years of age (as per the trial enrolment criteria); however, there were some issues that required clarification. In the first instance, in order to obtain this positive result, favourable outcome had to be redefined. Previously, in the stroke literature the mRS had been used to assess outcome, and this was usually dichotomised into favourable (mRS 0–3) or unfavourable (mRS 4–6). The implication being that the aim of surgery is to achieve an outcome which is felt to be acceptable to the person on whom the procedure is being performed. Notwithstanding some limitations of the mRS, the fundamental issue that denotes a favourable outcome is that a person has a degree of independence.

In the pooled analysis, favourable was reclassified such that it included patients with a mRS of 4. This would therefore include patients who cannot walk unaided and cannot look after their bodily needs which is an outcome that has been regarded as unfavourable for many years. Indeed, closer examination of the data confirms that the increase in survival came almost directly at the expense of an increase in the number of patients with a mRS of 4 [6]. Amongst the survivors who were randomised to receive standard medical care, 75% (9 of 12) had a mRS of 3 or less, an outcome previously defined as favourable. A similar favourable outcome was only achieved in the 55% of the patients treated surgically (22 of 40). It would seem that the most likely outcome following medical therapy is either death or a favourable outcome (mRS 0–3), whereas surgery considerably increases the risk of survival with a mRS of 4 [6].

The DESTINY II trial investigated the use of decompressive hemicraniectomy in patients over 60 years of age. In a similar design to the previous European trials, patients were randomised within 48 h of presentation to either surgical decompression or standard medical therapy. The results of this study confirmed that surgical intervention significantly reduced mortality; however the conclusion that “hemicraniectomy increased survival without severe disability” requires clarification (Table 13.1).

Most of the survivors in the hemicraniectomy group were adjudged to be either a mRS of either 4 or 5 which would seem to confirm the results of the previous studies that an increase in survival comes at the expense of survival with dependency. However, the finding that 63% of the survivors in the hemicraniectomy group gave retrospective consent to treatment would appear to support ongoing use of the procedure in this age group because it is difficult to argue that it is not on a patient’s best interest to have surgery if they are able to state that they are satisfied with their outcome and they would do the same again.

Yet closer examination of results shows that amongst the 27 survivors in the hemicraniectomy group, only 11 could adequately answer this question. The remaining 16 patients had to have a surrogate response from their next of kin because they themselves could not adequately answer the question because they had either severe aphasia or neuropsychological deficits. Given that only 7% (or three patients) achieved a mRS of 3, the remaining 24 patients had either mRS of 4 or 5. Assuming those patients with a mRS of 3 responded positively to the question regarding retrospective consent, amongst the remaining 24 patients, 16 patients could not walk without assistance, could not take care of their basic bodily needs and did have sufficient neurocognitive function such that they could answer a relatively simple question. Given what many informed commentators of sound mind would regard as an unacceptable outcome, these findings would encourage a reconsideration of the aforementioned conclusion [7].

Notwithstanding these observations, there is no doubt that the results of the trials provide good quality level 1 evidence on which to base clinical decision-making especially when considering patient selection and timing of surgery.

Realistically, in the emotionally charged atmosphere of an acute stroke, it would be very difficult to withhold therapy in a person under 60 years of age if there was at least some chance of survival with an acceptable level of disability, and the possibility of unacceptable dependency was acknowledged and accepted by those involved in making the decision. Treatment based on such reasoning can be justified even if the eventual outcome seems unacceptable to the injured party because risks and uncertainties are inevitable in all fields of medicine. It could also be argued given the alternative would be not to survive at all, a young person may quite reasonably be given the chance to “risk” survival with a mRS of 4 in the hope that they will either improve to achieve a mRS of 3 or learn to accept a level of disability that they might previously have deemed unacceptable.

For patients over 60 years of age, it must be acknowledged that there is only a very small chance that that person may achieve independence and there is a high chance that they may be left with very significant neurocognitive and neuropsychological deficits. Ideally, the acceptability or otherwise of this outcome for that particular individual must be fully explored prior to surgical intervention.

There is little doubt that certain individuals may feel that life is sacrosanct and worth preserving under any circumstances, and this may be based on certain religious, cultural or personal values. These individuals may also be willing to “run the risk” of survival with severe disability, in the hope that they may achieve a good functional outcome. They might also want the opportunity to adapt and learn to live with a level of disability that they and many others, perhaps, might previously have thought to be unacceptable. Whilst these views may fall outside what is deemed acceptable to the majority, where possible, these views should be acknowledged and acted upon. For this group of patient’s surgical intervention, following the development of “malignant” cerebral infarction is entirely reasonable.

The final group of patients will be those patients who have previously expressed a view (either previously voiced or documented) that they would not want to survive with severe disability. In these circumstances the surgeon cannot reasonably assume that they would be able to obtain consent for the operation and, if they did proceed, would have to justify acting on their own judgement against a properly considered assessment of the wishes of the patient.