Medication non-adherence assessed using liquid chromatography tandem mass spectrometry in ambulatory patients with heart failure referred for device therapy assessment

Background Medication nonadherence in heart failure (HF) is common, yet reported rates vary widely and are often based on self-report rather than objective measurement. Patients undergoing assessment for device therapy are a high-risk group in whom perceived failure of medical therapy may be attributable to unrecognised medication non-adherence. The prevalence of medication non-adherence in this group is not well studied. Aims To determine the prevalence and clinical correlates of medication non-adherence in ambulatory patients with HF referred for device therapy assessment. Methods We studied 58 patients with HF referred for outpatient assessment for device therapy. Plasma samples were obtained and analysed by liquid chromatography–tandem mass spectrometry (LC–MS) for prescribed cardiovascular medications, including renin–angiotensin–aldosterone system inhibitors (RAASi), beta-blockers (BB), mineralocorticoid receptor antagonists (MRA), diuretics, digoxin, and anticoagulants. Medication non-adherence was defined as non-detection of ≥1 prescribed medication in the plasma sample. Logistic regression was used to identify factors associated with non-adherence. Results Median age was 70 years (IQR 57–77); 67% male and 98% White. Median systolic blood pressure (SBP) and heart rate (HR) were 121 mmHg (IQR 111–138) and 69 bpm (IQR 61–80), respectively. Coronary artery disease was present in 21% (n=12), chronic kidney disease in 24% (n=14), and atrial fibrillation in 69% (n=40). Thirty-six patients (62%) were prescribed ≥5 medications. Following clinic assessment, 32 patients (55%) received device therapy. Overall, 22 (38%) patients were non-adherent to ≥1 prescribed medication. Among patients who received device therapy, half (n=15, 47%) were non-adherent to ≥1 medication. Rate of non-adherence varied by medication class, with the highest rates observed in digoxin (64%) and diuretics (37%). (Figure) Compared with adherent patients, those who were non-adherent had lower SBP (median 116 vs 125 mmHg, p=0.04), higher HR (median 80 vs 66 bpm; p=0.04), higher pill burden (median 7 vs 4 medications; p<0.001), and worse renal function (median creatinine 102 vs 86 µmol/L; p=0.04). In multivariable analysis adjusted for age and sex, lower SBP (per 10 mmHg decrease: OR=1.41, 95% CI 1.01–1.96; p=0.04), higher HR (per 10 bpm increase: OR 1.53, 95% CI 1.03–2.27; p=0.03), number of comorbidities (OR 2.29, 95% CI 1.14–4.60; p=0.02), and total number of prescribed medications (OR 1.65, 95% CI 1.20–2.27; p=0.002) were associated with non-adherence. Conclusion In ambulatory patients with HF referred for device therapy assessment, over one in three had biochemical evidence of non-adherence on single-sample plasma testing. Greater comorbidity and higher pill burden were independently associated with non-adherence. Incorporating drug-level testing may support optimisation of medical therapy prior to device therapy decisions.