Evaluation of the Potential Acetylcholinesterase Inhibitor‐Induced Rhinorrhea Prescribing Cascade
Year: 2019
Type: article
Study design: Observational Study
Abstract: The most commonly used medications for patients with dementia are acetylcholinesterase inhibitors (AChEIs; donepezil, rivastigmine, galantamine).1 Use of these medications are often limited by adverse events (AEs) including nausea, vomiting, diarrhea, incontinence, hypersalivation, sweating, lacrimation, rhinorrhea, and bronchorrhea.1-3 In particular, rhinorrhea is not a well-recognized AE of AChEI and was only previously documented in a meta-analysis of randomized controlled trials of donepezil.4 This AE may result in a prescribing cascade (PC) in which AChEI-induced rhinorrhea results in the use of a rhinorrhea medication rather than discontinuation or a dose reduction of AChEI.5 The AChEI-induced rhinorrhea PC has not been described beyond a case report.6 We aimed to assess this PC using a sample of patient visits from the United States. We used the 2014 and 2015 National Ambulatory Medical Care Survey (NAMCS) data that comprises medications from providers' notes including prescription and over-the-counter (OTC) products. Patient visits included those aged 55 years and older and treated with an antidementia medication (AChEI or memantine, an N-methyl-D-aspartate [NMDA] antagonist). We compared AChEI users (AChEI with or without memantine) with non-AChEI users (memantine only) because memantine is not associated with rhinorrhea.7 We assessed the co-use of rhinorrhea medications (RMs; ie, medications used to treat rhinorrhea) that served as markers for a potential AChEI-induced rhinorrhea PC. RMs included first-generation histamine1 (H1)-blockers, second-generation H1 blockers, nasal anticholinergics, nasal glucocorticoids, and montelukast, a leukotriene antagonist (full list available on request).8 We adjusted for age, sex, race, geographic region, year, comorbidities, and number of concomitant medications excluding RMs.9 Data were analyzed using the sampling visit weights that were used to produce national estimates and account for survey clustering. We compared the percentage of visits in which RMs were used among AChEI users vs non-AChEI users. Weighted odds ratios (ORs) and 95% confidence intervals (CIs) were used to calculate univariate and multivariable analyses. We also estimated the percentage of patient visits impacted by the AChEI-induced rhinorrhea PC by subtracting the percentages of AChEI users from non-AChEI users who co-used RMs. There were an estimated 14 743 000 (95% CI = 11 493 000-17 993 000) patient visits in which a patient aged 55 years or older was taking an antidementia medication (using weighted estimates). Among antidementia medication user visits, 85.6% (n = 12 606 000) were taking an AChEI (with donepezil the most common [n = 10 813 000]). Memantine was used in 35.3% of the visits (n = 5 203 000). Characteristics of AChEI users vs non-AChEI users are described in Table 1. Among visits in antidementia medication users, 18.1% (n = 2 665 000) co-used an RM. Second-generation H1 blockers were the most commonly used medication class (58.0% [n = 1 546 000]). Among antidementia medication patient visits, 20.6% (n = 12 606 000) of AChEI users and 3.5% (n = 75 000) of non-AChEI users utilized an RM. AChEI users were 7 times more likely to co-use an RM compared with non-AChEI users (OR = 7.16; 95% CI = 2.25-22.73). After adjusting for confounders, AChEI users were 4.7 times more likely to co-use an RM (OR = 4.70; 95% CI = 1.53-14.43) compared with non-AChEI users. The difference in RMs between AChEI users and non-AChEI users, used to estimate the number of AChEI users with AChEI-induced rhinorrhea PC, was 17.1% (n = 2 155 626). In this study using NAMCS data, we identified further evidence of the existence of the AChEI-induced rhinorrhea PC. Patients on AChEI were 4 times more likely to co-use an RM compared with no AChEI use, suggesting 17% of patient visits involving an AChEI may be experiencing the PC. The AChEI-induced rhinorrhea PC is considered problematic because it can result in negative consequences in older adults. As noted in the case report, the patient experienced clinically significant worsened cognition after using diphenhydramine as part of the AChEI-induced rhinorrhea PC.6 This particular cascade is also troublesome because symptoms may be self-treated using OTC products, in which OTC products may not be consistently identified during a clinic visit.10 Like similar PC research,9 the increasing number of concomitant medications was associated with the PC, thus further confirming that polypharmacy is associated with PCs. Our study has limitations. We used cross-sectional data and therefore could not confirm temporality of the PC. We were unable to differentiate AChEI doses, which are not reported in NAMCS, because rhinorrhea risk appears to be dose dependent. Self-administered OTC products may not be reported to the provider and therefore not captured in NAMCS; however, this is a likely differential between AChEI users and non-AChEI users. In conclusion, our findings support the presence of the AChEI-induced rhinorrhea PC. Our results are particularly troubling for older adults with dementia in whom the potentially unnecessary addition of an RM cannot only contribute to polypharmacy but result in poor outcomes such as worsening cognition. No funding. The authors have declared no conflict of interest for this article. Scott Martin Vouri performed the data analysis. Scott Martin Vouri and Marie-Christin Possinger designed the study, acquired the data, and drafted the article. All the authors critically revised, commented, and approved the final article. None.
DOI: 10.1111/jgs.16224
Language: English
Subfield: Geriatrics and Gerontology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2980425144