5 days ago
Drug Combinations for CVDs Tied to Bullous Pemphigoid Risk
TOPLINE:
A case-control study revealed that combinations of drugs for cardiovascular diseases and hypertension were frequently prescribed before a diagnosis of bullous pemphigoid (BP), but the risk associated with combinations did not exceed that associated with individual agents. The most common drug combinations included angiotensin-converting enzyme (ACE) inhibitors with statins and antiplatelets with statins.
METHODOLOGY:
Researchers conducted a nested case-control study using healthcare records from the Clinical Practice Research Datalink between 1998 and 2021 and analysed 16,844 BP cases and 79,493 age- and sex-matched control individuals having no BP diagnosis at the index date (the first date a BP diagnosis code was recorded).
Association rule mining (ARM) identified the 10 most common drug class or active substance pairs prescribed to cases or control individuals on the same day and within 6 months before the index date. In the sensitivity analysis, researchers identified medication pairs prescribed within 30 days of each other and during the 6 months preceding the index date.
Researchers quantified how often two drugs are co-prescribed compared with their independent prescribing by calculating a lift. They then derived the fold change (FC) as the ratio of a lift in cases vs control individuals.
The analysis included multivariable conditional logistic regression to estimate the risk for BP following drug combinations and their constituent drugs.
TAKEAWAY:
The most frequent drug combinations associated with an increased risk for BP were ACE inhibitors-statins (FC of the lifts in the main analysis vs sensitivity analysis: 1.31 vs 1.18), antiplatelets-statins (1.23 vs 1.11), proton pump inhibitors (PPI)-antiplatelets (1.22 vs 1.14), PPI-statins (1.22 vs 1.14), and ACE inhibitors-antiplatelets (1.20 vs 1.09).
For drug substances, combinations with a greater lift in BP cases were simvastatin-ramipril (FC, 1.30), simvastatin-aspirin (FC, 1.21), and ramipril-aspirin (FC, 1.19).
After adjusting for BP-associated drugs, the Charlson Comorbidity Index, and relevant confounders, the increased risk remained significant for these drug class combinations: antiplatelets-statins (odds ratio [OR], 1.20), ACE inhibitors-statins (OR, 1.16), PPI-statins (OR, 1.22), ACE inhibitors-antiplatelets (OR, 1.26), and PPI-antiplatelets (OR, 1.43; P < .001 for all).
The risk for BP associated with these frequently prescribed drug combinations was lower than the risk linked to each constituent drug at both class and substance levels.
In both main and sensitivity analyses, patients who developed BP were more likely than control individuals to have received combinations of cardiovascular or antihypertensive drugs before diagnosis.
IN PRACTICE:
"The ARM algorithm exploratory analysis identified the most commonly prescribed drug combinations prior to BP. Logistic regression confirmed drug combinations for CVDs [cardiovascular diseases] or hypertension associated with increased BP risk," the authors wrote. "The increased BP risk following reported combinations was modest and was not greater than their constituent drugs. Given that the number of patients with BP is low, we do not suggest avoiding the reported drugs but instead being on the lookout for any skin reactions following treatments for CVDs or hypertension," they concluded.
SOURCE:
This study was led by Mikolaj Swiderski, University of Nottingham, Nottingham, England. It was published online on August 06, 2025, in Clinical and Experimental Dermatology.
LIMITATIONS:
The ARM algorithm considered only the frequency of prescriptions to obtain drug combinations. Additionally, the algorithm demonstrated limited clinical value, linking only half of the inferred drug class combinations with BP and failing to capture the sequence or precise timing of prescriptions. It also lacked dosage and treatment duration data, and as an exploratory tool, ARM could not establish causal relationships between drug exposures and the risk for BP.
DISCLOSURES:
This research was supported by the National Institute for Health and Care Research grant via the Research for Patient Benefit Programme. Swiderski reported receiving salary funding from this grant. Another author reported receiving salary funding from King's College London, University of Nottingham, and the National Institute for Health and Care Research East Midlands scholarship scheme. Additional disclosures are noted in the original article.
This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.
