Postoperative delirium (POD) is a common complication in older adult patients after surgery. A patient's preoperative anticholinergic (AC) burden is a potentially modifiable risk factor for POD. As the influence of the drug dose remains unknown, we aimed to compare three AC burden scores in relation to POD, two of which were dose-related. This retrospective cohort study (03/22-10/22) included orthopaedic and trauma surgery patients > 65 years. POD was assessed using the four A's test (4AT), delirium diagnosis, and chart review. The AC burden was determined using the non-dose-related German Anticholinergic Burden score (GerACB), an extension of the dose-related Muscarinic Acetylcholinergic Receptor ANTagonist Exposure scale (extMARANTE), and the dose-related German Drug Burden Index (GerDBI). Multivariable logistic regression analysis determined the association between the preoperative AC burden and POD. Scores were compared using kappa statistics, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). POD was observed in 71 of 385 patients (18.4%). For all three scores, a high AC burden was significantly associated with POD after adjusting for age, sex, dementia, preoperative physical status, and number of prescribed drugs (p < 0.001). The overall agreement among the burden classifications was substantial (no POD: κ = 0.645, POD: κ = 0.632). The GerACB had the lowest sensitivity with 23.9% (extMARANTE: 42.3%, GerDBI: 40.8%), but the highest PPV with 48.6% (extMARANTE: 38.5%, GerDBI: 43.3%). Both dose-related and non-dose-related AC burden scores have limited sensitivity and modest PPV for screening a patient's medication for POD. However, given the additional effort required for dose consideration, the non-dose-related GerACB remains sufficient in clinical practice, with the lowest sensitivity but highest PPV.
Read full abstract