More than 90% of patients reached the neurosurgical center within 12 hours of their first hospital admission after SAH; 70% of all aneurysms were clipped or coils were inserted within 24 hours of the first hospital admission. Given the protocol, only one rebleed occurred later than 24 hours after the first hospital admission. Despite this strong emphasis on early intervention, however, a cluster of 27 very early rebleeds still occurred in the control group within hours of randomization into the study, and 13 of these patients died. In the tranexamic acid group, six patients rebled and two died. A reduction in the rebleeding rate from 10.8 to 2.4% and an 80% reduction in the mortality rate from early rebleeding with tranexamic acid treatment can therefore be inferred. Favorable outcome according to the GOS increased from 70.5 to 74.8%. According to TCD measurements and clinical findings, there were no indications of increased risk of either ischemic clinical manifestations or vasospasm that could be linked to tranexamic acid treatment. Neurosurgical guidelines for aneurysm rupture should extend also into the preneurosurgical phase to guarantee protection from ultraearly rebleeds. Currently available antifibrinolytic drugs can provide such protection, and at low cost. The number of potentially saved lives exceeds those lost to vasospasm.

The results obtained in this series closely reflect the overall management results of this disease and support the conclusion that surgical complications causing a poor outcome can be estimated on a large population-based scale. Intraoperative aneurysm rupture was the most common and most devastating technical complication that occurred. Support was found for a more liberal use of temporary clips early during dissection, regardless of the experience of the surgeon. Temporary regional interruption of arterial blood flow should be a routine method for aneurysm surgery on an everyday basis. A random occurrence of difficult intraoperative problems was clearly shown, and this factor of unpredictability, which is present in any preoperative assessment of risk, strengthens the case for recommending neuroprotection as a routine adjunct to virtually every aneurysm operation, regardless of the surgeon's experience.

The impact of warning leaks on management results in patients with aneurysmal subarachnoid hemorrhage (SAH) was evaluated in this prospective study. In a consecutive series of 422 patients with aneurysmal SAH, 84 patients (19.9%) had an episode suggesting a warning leak; 34 (40.5%) of these patients were seen by a physician without the condition being recognized. The warning leak occurred less than 2 weeks before a major SAH in 75% of the patients. A good outcome was experienced by 53.6% of patients who had a warning leak versus 63.3% of those who had no warning leak. In a subgroup of patients who had an interval of 3 days or less from warning leak to SAH, only 36.4% had a good outcome. The proportion of patients in good neurological condition (Hunt and Hess Grades I and II) who had a good outcome was 88.1% in the group with no warning leak versus 53.6% in the group whose SAH was preceded by a warning leak. A difference of 35% between these two groups reflects the impact of an undiagnosed warning leak on patient outcome, based on the assumption that patients with a warning leak had clinical conditions no worse than Hunt and Hess Grade II at the time of the episode. In the subgroup of patients with the short interval between warning leak and SAH, the difference was almost 52%. The difference in outcome also reflects the potential improvement in outcome that can be achieved by a correct diagnosis of the warning leak. If the correct diagnosis is made in patients seeking medical attention due to a warning leak, favorable outcomes in the overall management of aneurysmal SAH are estimated to increase by 2.8%. An active diagnostic attitude toward patients experiencing a sudden and severe headache is warranted as it offers a means of improving overall outcome in patients with SAH.

Objective: An important safety issue with GH replacement therapy (GHRT) in hypopituitary patients with a history of a pituitary adenoma is the risk for tumour recurrence or enlargement. Design: Case-control study. Subjects and methods: We studied tumour progression rate in 121 patients with hypopituitarism on the basis of non-functioning pituitary adenomas (NFPA) receiving long-term GHRT. A group of 114 NFPA patients not receiving GHRT who were matched in terms of duration of follow-up, gender, age, age at diagnosis and radiotherapy status were used as a control population. The average duration of GHRT was 10G4 years (range 2-17).

Objective: Most patients who have been treated for craniopharyngioma (CP) are GH deficient (GHD). GH replacement therapy (GHRT) may stimulate tumour regrowth; and one of the concerns with long-term GHRT is the risk of tumour progression. Therefore, the objective was to study tumour progression in CP patients on long-term GHRT. Design: Case-control study. Patients and methods:The criteria for inclusion of cases were: i) GHD caused by CP; ii) GHRT O3 years; and iii) regular imaging. This resulted in 56 patients (mean age at diagnosis 25G16 years) with a mean duration of GHRT of 13.6G5.0 years. As controls, 70 CP patients who had not received GHRT were sampled with regard to follow-up, gender, age at diagnosis and initial radiation therapy (RT). Results: The 10-year tumour progression-free survival rate (PFSR) for the entire population was 72%. There was an association (hazard ratio, P value) between PFSR and initial RT (0.13, !0.001) and residual tumour (3.2, !0.001). The 10-year PFSR was 88% for the GHRT group and 57% for the control group. Substitution with GHRT resulted in the following associations to PFSR: GHRT (0.57, 0.17), initial RT (0.16, !0.001), residual tumour (2.6, !0.01) and gender (0.57, 0.10). Adjusted for these factors, the 10-year PFSR was 85% for the GHRT group and 65% for the control group. Conclusions: In patients with CP, the most important prognostic factors for the PFSR were initial RT and residual tumour after initial treatment. Long-term GHRT did not affect the PFSR in patients with CP.