CT diagnosis and outcome of primary brain tumours in the elderly: a cohort study

BACKGROUND: The incidence of primary brain tumours (PBT) increases with age. Survival outcome depends on the treatment modality and histological type of the tumour.

OBJECTIVES: To compare the survival outcome between those who had brain biopsy and those who did not among those who had PBT diagnosed by computerised tomography (CT).

METHODS: We analysed data from 82 elderly patients who were admitted to a UK centre with close links to a neurosurgical unit.

RESULTS: Age range was 62-99 years (median 74 years); 49 males and 33 females. 44 cases (54%) had brain biopsies of which 2 (5%) were not PBT. Both cases were malignant conditions, 1 case of metastasis and 1 leiomyosarcoma. Of 42 PBT, 34 (77%) were malignant gliomas. The remaining 8 cases were 4 astrocytomas, 2 meningiomas, 1 cerebellar tumour and 1 cerebral lymphoma. Initial clinical misdiagnosis of cerebrovascular disease (CVA) or transient ischaemic attack (TIA) was very common, 70% in our cohort. Those who had brain biopsy were younger (median age 72, range 62-81) compared to those who did not have brain biopsy (median age 77, range 65-91). A survival analysis for those who died comparing biopsy-proven malignant tumours (combined gliomas and astrocytomas (n = 37)) and those who did not have a biopsy (n = 37) showed that the latter had a shorter median time to receive definitive treatment after diagnosis, 8.0 days (95% CI; 5.4, 8.6) (n = 20) compared to 40.0 days (95% CI; 32.5, 47.5) (n = 26) but shorter median length of survival, 47.0 days (95% CI; 32.3, 61.7) compared to 81.0 days (95% CI; 66.7, 95.3) for those who had brain biopsy. Subgroup analysis of malignant gliomas between patients aged < or =70 years and those >70 years showed no significant difference in median length of survival, 74.0 days (95% CI; 54.0, 94.0) vs. 85.0 days (95% CI; 59.0, 111.0), respectively.

CONCLUSIONS: Our findings suggest that the older elderly are less likely to have aggressive therapy, and more likely to be diagnosed initially as cerebrovascular event (CVA/TIA). However, when they were given similar definitive treatment, their survival is comparable with their younger counterparts and therefore, age alone should not contraindicate radical treatment. Elderly patients should be selected for radical treatment on existing criteria, but age itself should not preclude radical treatment where it is otherwise appropriate to offer it.