Association of preoperative depression and survival after resection of malignant brain astrocytoma.

BACKGROUND: Clinical depression has been shown to negatively influence the morbidity and mortality of multiple disease states. It remains unclear if clinical depression affects survival after surgical management of malignant brain astrocytoma. We set out to determine whether patients with a diagnosis of clinical depression before surgery experienced decreased survival independent of treatment modality or degree of disability. METHODS: One thousand fifty-two patients undergoing surgical management for malignant brain astrocytoma (WHO grade 3 or 4) performed at a single institution from 1995 to 2006 were retrospectively reviewed. The independent association of depression prior to surgery and subsequent survival was assessed via multivariate proportional hazards regression analysis. RESULTS: Surgical management consisted of primary resection in 605 (58%) patients, secondary resection in 410 (39%), and biopsy in 37 patients (3.5%). Pathology was WHO grade IV in 829 (79%) and grade III in 223 (21%). Forty-nine patients (5%) carried the diagnosis of depression at the time of surgery. Mean age and KPS on admission was 51 +/- 16 and 80 +/- 10 years, respectively. Two hundred ninety patients (28%) received Gliadel (BCNU MGI Pharma, Inc., Bloomington, MN, USA) wafer implantation and 274 (26%) received postoperative temozolomide (concomitant in 102, delayed adjuvant in 172 patients). Subsequent resection was performed at the time of recurrence in 135 (13%) patients a mean of 10 +/- 6 months after surgery. Adjusting for all variables associated with survival in this model, age (P < .001), KPS (P < .001), WHO grade III vs IV (P < .001), primary versus secondary resection (P < .001), gross-total resection (P < .001), Gliadel wafer implantation (P = .048), postoperative temozolomide therapy (P < .001), and subsequent resection at time of recurrence (P < .001); preoperative depression was independently associated with decreased survival (relative risk [95% CI]: 1.41 [1.1-1.96], P < .05). The difference in percent survival between the depression and nondepression cohorts was most notable at 12 months (15% vs 41%) and 20 months (0% vs 21%) after surgery. CONCLUSION: In our experience, patients who are actively depressed at the time of surgery were associated with decreased survival after surgical management of malignant astrocytoma, independent of degree of disability, tumor grade, or subsequent treatment modalities. In our opinion, the presence of an association between preoperative depression and survival warrants further investigation.

Duke Scholars

Author Khoi Duc Than Neurosurgery