Reversibility of delirium in Ill-hospitalized cancer patients: Does underlying etiology matter?

Background The objective of this study was to explore the underlying etiologies associated with the resolution and improvement of delirium in ill-hospitalized cancer patients. Methods We conducted a secondary analysis of a multicenter, prospective, observational study to estimate the effectiveness of pharmacotherapy for delirium. Participants were cancer patients with delirium. We assessed the Delirium Rating Scale, Revised-98 (DRS-R98) severity scale score at baseline and three days after pharmacotherapy initiation. Delirium resolution was defined as a DRS-R98 severity scale score <= 9, and improvement was defined as >= 50% reduction at Day 3. Results We enrolled 566 patients (491 patients had performance status of 3 or 4). The resolution and improvement rates in all patients were 22.6% and 19.3%, respectively. Univariate analysis determined that nonrespiratory infection (OR 2.18, 95% CI 1.38-3.45) was significantly associated with greater resolution, while dehydration (0.40, 0.19-0.87), organic damage to the central nervous system (CNS) (0.32, 0.43-0.72), hypoxia (0.25, 0.12-0.52), and hyponatremia (0.34, 0.12-0.97) were significantly associated with no resolution. Potential causes associated with delirium improvement were nonrespiratory infection (1.93, 1.19-3.13), organic damage to the CNS (0.40, 0.18-1.90), and hypoxia (0.32, 0.16-0.65). After multivariate analysis, dehydration (0.34, 0.15-0.76), organic damage to the CNS (0.25, 0.10-0.60), and hypoxia (0.29, 0.14-0.61) were significantly associated with no resolution. Conclusions Delirium caused by nonrespiratory infection may be reversible, while delirium associated with dehydration, organic damage to the CNS, hypoxia, or hyponatremia seems to be irreversible in ill-hospitalized cancer patients.