PURPOSE: Few studies looking for an effective biomarker to predict delirium have been performed. This study was designed to investigate whether a change in inflammatory status, indicated by blood natural killer (NK) cell activity, predicts delirium. METHODS: This prospective study, performed in 4 university and 1 general hospital from September, 2011 to October, 2012, included 29 patients. Eligible patients were 65-89years old, newly and emergently admitted. Patients were assessed daily, up to 7days, for occurrence of DSM-IV-defined delirium. The main outcome measure was change in blood NK cell activity between the first and second mornings after admission. RESULTS: The mean change in blood NK cell activity on the second morning, compared to the first morning, in patients developing delirium (n=9) was significantly greater than that in patients without delirium (n=20) (6.0% [SD 8.4] vs. -1.4% [9.0], respectively, t=2.10, P=0.045). Significant difference between the groups was still found after adjusting for age, the history of previous delirium, and the Clinical Dementia Rating score (F=6.63, P=0.017). Of note is that 8 of 9 (89%) patients developing delirium had increased blood NK cell activity, as did only 8 of 20 (40%) patients without delirium, giving measurement of this parameter, for distinguishing the two groups, a sensitivity of 89%, specificity 60%, positive predictive value 50%, negative predictive value 92%, positive likelihood ratio 2.22, and negative likelihood ratio 0.19. When combining this predictor with another predictor, a Delirium Rating Scale-Revised-98 severity score of 5 or more at baseline, positive and negative likelihood ratios were 7.80 and 0.24, respectively. CONCLUSION: Increase in blood NK cell activity may be associated with developing delirium. Further studies including larger numbers of patients are needed to justify the preventive use of drugs for patients meeting criteria for both predictors.
PURPOSE: Few studies looking for an effective biomarker to predict delirium have been performed. This study was designed to investigate whether a change in inflammatory status, indicated by blood natural killer (NK) cell activity, predicts delirium. METHODS: This prospective study, performed in 4 university and 1 general hospital from September, 2011 to October, 2012, included 29 patients. Eligible patients were 65-89years old, newly and emergently admitted. Patients were assessed daily, up to 7days, for occurrence of DSM-IV-defined delirium. The main outcome measure was change in blood NK cell activity between the first and second mornings after admission. RESULTS: The mean change in blood NK cell activity on the second morning, compared to the first morning, in patients developing delirium (n=9) was significantly greater than that in patients without delirium (n=20) (6.0% [SD 8.4] vs. -1.4% [9.0], respectively, t=2.10, P=0.045). Significant difference between the groups was still found after adjusting for age, the history of previous delirium, and the Clinical Dementia Rating score (F=6.63, P=0.017). Of note is that 8 of 9 (89%) patients developing delirium had increased blood NK cell activity, as did only 8 of 20 (40%) patients without delirium, giving measurement of this parameter, for distinguishing the two groups, a sensitivity of 89%, specificity 60%, positive predictive value 50%, negative predictive value 92%, positive likelihood ratio 2.22, and negative likelihood ratio 0.19. When combining this predictor with another predictor, a Delirium Rating Scale-Revised-98 severity score of 5 or more at baseline, positive and negative likelihood ratios were 7.80 and 0.24, respectively. CONCLUSION: Increase in blood NK cell activity may be associated with developing delirium. Further studies including larger numbers of patients are needed to justify the preventive use of drugs for patients meeting criteria for both predictors.