ObjectiveAngiotensin-converting enzyme inhibitors are widely prescribed for cardiovascular disease and are a well-recognized cause of cough. However, cough in patients treated with angiotensin-converting enzyme inhibitor may also reflect underlying pulmonary disease. This study compared the clinical, laboratory, and medication-related characteristics of angiotensin-converting enzyme inhibitor users presenting with pulmonary and non-pulmonary cough.MethodsThis retrospective single-center study included 199 angiotensin-converting enzyme inhibitor-treated patients who presented with cough to the Chest Diseases Outpatient Clinic of Giresun Training and Research Hospital between October 2021 and January 2025. Patients were classified as having cough with pulmonary or non-pulmonary causes according to available clinical, laboratory, and radiological findings. Non-pulmonary cough was not considered synonymous with confirmed angiotensin-converting enzyme inhibitor-induced cough; however, it indicated the absence of identifiable pulmonary pathology based on retrospective data. Demographic, clinical, laboratory, and angiotensin-converting enzyme inhibitor subtype characteristics were compared.ResultsAmong 199 patients, 153 (76.9%) had pulmonary cough and 46 (23.1%) had non-pulmonary cough. Cough duration was longer (p = 0.044) and dyspnea was more frequent (54.2% vs. 13.0%, p < 0.001) in the pulmonary group. Perindopril use was more common in the non-pulmonary group, whereas ramipril use was more common among patients with pulmonary cough (p = 0.015). In a post hoc analysis limited to patients with documented angiotensin-converting enzyme inhibitor discontinuation and evaluable follow-up notes (n = 78), the overall distribution of cough-response categories differed between the groups (Fisher-Freeman-Halton exact p = 0.039), with complete resolution observed in 33.3% of the patients in the non-pulmonary group and 7.9% patients in the pulmonary group. C-reactive protein levels were higher in the pulmonary group (10.87 vs. 5.67 mg/L, p = 0.010), whereas other hematological parameters did not differ significantly.ConclusionsIn the angiotensin-converting enzyme inhibitor-treated patients presenting with cough, shorter symptom duration, absence of dyspnea, and lower C-reactive protein levels may be more compatible with non-pulmonary or presumed angiotensin-converting enzyme inhibitor-related cough, whereas prolonged cough, dyspnea, and elevated C-reactive protein levels may support the consideration of a pulmonary pathology. These findings are hypothesis-generating and require prospective validation.

Background/Objectives: The global rise in obesity-related hypertension, metabolic syndrome, and chronic inflammation calls for a precise characterization of the safety profiles of first-line therapies. While amlodipine is considered metabolically neutral, its real-world impact on dysglycemia and inflammatory biomarkers remains incompletely defined. This study aims to characterize the metabolic and inflammatory adverse drug reaction profile of amlodipine, using the EudraVigilance database. Methods: Descriptive and disproportionality analyses were performed on 41,872 Individual Case Safety Reports recorded prior to 17 May 2026. Amlodipine was compared against major antihypertensive classes (beta-blockers, angiotensin-converting enzyme (ACE) inhibitors, sartans, and diuretics), calculating reporting odds ratios (ROR) and 95% confidence intervals. Results: "Hyperglycaemia" could be considered a safety signal for amlodipine, compared to beta-blockers (e.g., bisoprolol-ROR: 2.56), ACE inhibitors (e.g., ramipril-ROR: 2.82), and sartans (e.g., candesartan-ROR: 3.85). "Metabolic syndrome" was reported for amlodipine with a lower probability than for hydrochlorothiazide (ROR: 0.35). Inflammatory signals (e.g., increased C-reactive protein) appeared less frequently for amlodipine than for certain renin-angiotensin-aldosterone system inhibitors (e.g., perindopril-ROR: 0.53). Conclusions: The disproportionality analysis identified relatively lower reporting frequencies for several metabolic and inflammatory adverse drug reactions, compared with selected antihypertensive agents. These findings represent pharmacovigilance signals that warrant further investigation in analytical epidemiological and clinical studies. Although hyperglycemia was reported disproportionally relative to several comparator drugs, reports of T2DM were less frequently reported for amlodipine. However, these observations should not be interpreted as evidence of differences in clinical risk, because disproportionality analyses cannot establish incidence or causality. Reports of inflammation likely reflect patient comorbidities rather than a direct drug effect. These findings demonstrate that post-marketing surveillance remains essential, even when accounting for the methodological limitations of spontaneous reporting.

Limited information indicates that only low levels of perindopril and its active metabolite are found in breastmilk, which is consistent with other drugs in this class. Amounts ingested by the infant are small and would not be expected to cause any adverse effects in breastfed infants.