Abstract
Background: Although previous studies have established the association of medications with anticholinergic adverse effects and xerostomia, anticholinergic burden and xerostomia in critical care settings are poorly characterized. The objective of this study was to determine the impact of medication burdens associated with anticholinergic adverse effects, particularly the occurrence of xerostomia (dry mouth) in a critical care setting. In addition, this study explored the correlation between the timing of the first instance of xerostomia and the administration timing of medication known to have anticholinergic adverse effects.
Methods: A retrospective case-control study was used with the MIMIC (Medical Information Mart for Intensive Care) III database. The MIMIC-III clinical database is a publicly available, deidentified, health-related database with more than 40 000 patients in critical care units from 2001 to 2012. Cases of xerostomia (n = 1344) were selected from clinical notes reporting "dry mouth," "xerostomia," or evidence of pharmacological treatment for xerostomia; control (n = 4032) was selected using the propensity analysis with 1:3 matching on covariates (eg, age, sex, race, ethnicity, and length of stay). The anticholinergic burden was quantified as the cumulative effect of anticholinergic activities using the Anticholinergic Burden Scale.
Results: Anticholinergic burden significantly differed between xerostomia patients and control subjects (P = .04). The length of stay was a statistically significant factor in xerostomia. The probability of developing the symptom of xerostomia within 24 hours was .95 (95%) for patients of xerostomia.
Conclusions: Anticholinergic Burden Scale is associated with xerostomia in the critical care setting, particularly within 24 hours after admission. It is crucial to carefully evaluate alternative options for medications that may have potential anticholinergic adverse effects. This evaluation should include assessing the balance between the benefits and harms, considering the probability of withdrawal reactions, and prioritizing deprescribing whenever feasible within the initial 24-hour period.