Objective: The aim of this study was to analyze the clinical outcome of severely ill neurosurgical patients whose need for artificial life support was extended. We sought to determine whether these patients benefit from extended treatment both in life expectancy and quality of life. Furthermore, we evaluated the direct cost of the neurosurgical treatment.
Methods: The study group comprised a consecutive series of 346 neurosurgical patients in poor condition who were discharged from the intensive care unit but still in need of artificial respiratory support. The patients had various neurosurgical diagnoses and were treated between 2000 and 2003 at the Department of Neurosurgery, Helsinki University Central Hospital. We followed the outcome of these patients by specially formatted questionnaires 6 months and 1, 2, and 5 years after treatment. Their health-related quality of life was evaluated with EuroQol EQ-5D; quality-adjusted life years (QALY) gained with the treatment and the costs of a QALY were calculated.
Results: The median follow-up time was 5 years. The mortality rate was 27% at 30 days, 45% at 1 year, and 59% at 5 years after treatment. Of the patients, 20% had a good recovery (Glasgow Outcome Scale [GOS] scores 4 and 5), 18% had severe disability (GOS score 3), none was in a vegetative state (GOS score 2), 59% were dead (GOS score 1), and 3% were lost to follow-up. Of the survivors, 69% lived at home, 22% in a nursing home, 2% were in a hospital, and 7% were lost to follow-up. The median EQ-5D index value was lower than the median index value for the general population: 0.71 (25th percentile [Q1] 0.38 and 75th percentile [Q3] 0.85) versus 0.85 (Q1 0.73 and Q3 1.00). The median cost of the direct neurosurgical treatment per patient was 15,000 euros (25th percentile, 10,000 euros 75th percentile, 22,000 euros). Surviving patients gained a mean of 17 +/- 13 QALYs. The cost of 1 QALY was 2521 euros.
Conclusion: Prolonged intensive care unit and step-down unit treatment of critically ill neurosurgical patients seems to be clinically justified. Moreover, direct costs of neurosurgical treatment were reasonably low.