Evaluation of the association of anticholinergic burden and delirium in older hospitalised patients – A cohort study comparing 19 anticholinergic burden scales

Abstract

A recent review identified 19 anticholinergic burden scales (ABSs) but no study has yet compared the impact of all 19 ABSs on delirium. We evaluated whether a high anticholinergic burden as classified by each ABS is associated with incident delirium. Method:We performed a retrospective cohort study in a Swiss tertiary teaching hospital using data from 2015-2018. Included were patients aged ≥65, hospitalised ≥48 hours with no stay >24 hours in intensive care. Delirium was defined twofold:(i) ICD-10 or CAM and (ii) ICD-10 or CAM or DOSS. Patients' cumulative anticholinergic burden score, calculated within 24 hours after admission, was classified using a binary (<3: low, ≥3: high burden) and a categorical approach (0: no, 0.5-3: low, ≥3: high burden). Association was analysed using multivariable logistic regression.Results: Over 25 000 patients (mean age 77.9 ± 7.6 years) were included. Of these, (i) 864 (3.3%) and (ii) 2770 (11.0%) developed delirium. Depending on the evaluated ABS, 4-63% of the patients were exposed to at least one anticholinergic drug. Out of 19 ABSs, (i) 14 and (ii) 16 showed a significant association with the outcomes. A patient with a high anticholinergic burden score had odds ratios (ORs) of 1.21 (95% confidence interval [CI]: 1.03-1.42) to 2.63 (95% CI: 2.28-3.03) for incident delirium compared to those with low or no burden. Conclusion:A high anticholinergic burden within 24 hours after admission was significantly associated with incident delirium. Although prospective studies need to confirm these results, discontinuing or substituting drugs with a score of ≥3 at admission might be a targeted intervention to reduce incident delirium.

Comparative analysis of anticholinergic burden scales to explain iatrogenic cognitive impairment and self-reported side effects in the euthymic phase of bipolar disorders: results from the FACE-BD cohort

“…We replicated the lack of association between cognitive impairment in bipolar patients and Anticholinergic Drug Scale values reported by Eum et al (2017). Previous use of Chew's scale showed its association with cognitive impairment in elderly patients (Lampela et al, 2013;Lisibach et al, 2021), whereas the Anticholinergic Toxicity Scale has been shown to be strongly associated with delirium and mortality in elderly patients (Lisibach et al, 2022a(Lisibach et al, & 2022b, but no study has explored their association with cognitive performance for individuals with psychiatric disorders. Indeed, Chew's scale was designed based exclusively on in vitro measures of serum anticholinergic activity and the Anticholinergic Toxicity Scale was designed based on the affinity for muscarinic receptors deduced from the molecular structure of the drug: neither of the two scales were designed based on clinical observations, making our results quite unexpected.…”

Comparative analysis of anticholinergic burden scales to explain iatrogenic cognitive impairment and self-reported side effects in the euthymic phase of bipolar disorders: Results from the FACE-BD cohort

“…We replicated the lack of association between cognitive impairment in bipolar patients and Anticholinergic Drug Scale values reported by Eum et al (2017) . Previous use of Chew's scale showed its association with cognitive impairment in elderly patients ( Lampela et al, 2013 ;Lisibach et al, 2021 ), whereas the Anticholinergic Toxicity Scale has been shown to be strongly associated with delirium and mortality in elderly patients ( Lisibach et al, 2022a( Lisibach et al, & 2022b, but no study has explored their association with cognitive performance for individuals with psychiatric disorders. Indeed, Chew's scale was designed based exclusively on in vitro measures of serum anticholinergic activity, and the Anticholinergic Toxicity Scale was designed based on the affinity for muscarinic receptors deduced from the molecular structure of the drug: neither of the two scales was designed based on clinical observations, making our results quite unexpected.…”

Predicting delirium in older non-intensive care unit inpatients: development and validation of the DELIrium risK Tool (DELIKT)

“…We observed a delirium occurrence of 11%, which is at the lower end of the range found in the literature [ 1 ]. Although the DOSS is not considered to be a standard diagnostic tool, unlike the CAM, we included the DOSS in our delirium definition based on a previous sensitivity analysis [ 7 ], which did not show major differences in the effect size between the DOSS and the CAM. Additionally, including the DOSS might address the previously reported problem of delirium underdiagnosing [ 28 ].…”

“…The years 2015/2016 were used for development and 2017/2018 for external validation. The cohort selection criteria, outcome definition and predictors are summarised below and have been described elsewhere in more detail [ 7 ]. This study was undertaken per the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement [ 14 ].…”

“…The intake of ACH drugs is associated with a cumulative ACH burden due to age-related physiological changes, such as increased permeability of the blood–brain barrier, reduced renal and hepatic clearance and higher susceptibility and sensitivity to ACH medications [ 6 ]. In a recent publication, we found that a high cumulative ACH burden score of three points or more, measured with any of the published anticholinergic burden scales (ABSs) within the first 24 h of admission, is significantly associated with incident delirium in patients aged 65 years or older [ 7 ].…”