Alcohol consumption may be associated with postoperative delirium in the elderly: the PNDABLE study

Abstract

Abstract Objectives: The aim of this study was to reveal the relationship between alcohol consumption and Postoperative delirium (POD) in the elderly. Methods: A total of 284 patients were enrolled in this study (mean age was 69.38 (65-90) years and 7.1% were female). Those who had measurements of CSF Aβ40, Aβ42，P-tau, and tau protein and drinking characteristics were included from the Perioperative Neurocognitive Disorder And Biomarker Lifestyle (PNDABLE ) study. Mini-Mental State Examination (MMSE) was used to assess preoperative mental status of patients. POD was diagnosed using the Confusion Assessment Method (CAM) and assessed for severity using the Memorial Delirium Assessment Scale (MDAS). The two independent samples t-test was used for comparison between the two groups. Logistic regression analysis was utilized to explore the association of alcohol consumption with POD. What’s more, We also performed sensitivity analysis by adding corrected confounders and the results were almost unchanged. Linear regression analysis was used to study the relationship between alcohol consumption and CSF biomarkers. Mediation analyses with 10000 bootstrapped iterations were used to explore the mediation effects. A two-factor ANOVA was used to explore the impact of gender and alcohol consumption on CSF biomarkers. Finally, we constructed the receiver operating characteristic (ROC) curve and the nomogram model to evaluate the efficacy of alcohol consumption and CSF biomarkers in predicting POD. Result: The incidence of POD was 17.5% (44/252). Logistic regression showed that alcohol consumption (OR = 1.016, 95%CI 1.009-1.024, P < 0.001) is a risk factor for POD. In particular, alcohol consumption >24 g (heavy) is more likely to cause POD. What’s more, Aβ42 is a protective factor for POD (OR = 0.993, 95%CI 0.989-0.997, P < 0.05) and P-Tau was a risk factor for POD (OR = 1.093, 95%CI 1.022-1.168, P < 0.05). Linear regression analysis revealed that alcohol consumption was negatively associated with CSF Aβ42 (β = -0.638, P < 0.001) in POD patients. The two independent samples t-test showed that the non-postoperative delirium (NPOD) group had higher Aβ42 levels than the POD group; the POD group had more alcohol consumption than the NPOD group (P < 0.001). Mediation analyses show that alcohol consumption is likely to mediate POD through Aβ42 (proportion:14.21%) partially. ROC curve showed that alcohol consumption (AUC=0.904; P<0.001) exhibited a relative better discriminatory ability in POD prediction compared to Aβ42 (AUC=0.798; P<0.001). The calibration curve indicated good prediction of the nomogram(P=0.797). Conclusion: Alcohol consumption is a risk factor for POD (particularly for those with >24 g a day on average) in the elderly, and contributes to POD through the mediation of Aβ42.