Phd in INNOVATIVE BIOMEDICAL TECHNOLOGIES IN CLINICAL MEDICINE

Thesis title: Clinical Complexity in Atrial Fibrillation: from determinants to comprehensive and integrated management according to the ABC pathway

BACKGROUND: The epidemiology of atrial fibrillation (AF) is changing worldwide, and patients with AF are increasingly older and burdened with comorbidities; the interaction of different conditions and health determinants (such as age, frailty, multimorbidity and polypharmacy) entails the so-called “clinical complexity”, i.e. a clinical condition characterized by challenges in clinical management, complex health needs, and higher risk of adverse. Most studies have focused on the evaluation of single determinants of clinical complexity, although this approach may lack the comprehensiveness requested to investigate such condition. Moreover, the ‘Atrial fibrillation Better Care’ (ABC) pathway has been proposed to streamline a holistic or integrated care approach, to improve management and prognosis in patients with AF; the ABC pathway encompasses a specific focus on the management of comorbidities and associated risk factors, and has been found effective in improving outcomes in patients with AF, thus representing a promising approach to improve management and prognosis in clinically complex patients with AF. AIMS: Main aims of this project were 1) to assess clinical complexity in real-world patients with AF, according to levels and phenotypes of clinical complexity; and 2) to evaluate the effectiveness of the ABC pathway in improving outcomes in patients with AF, across different levels and phenotypes of clinical complexity. METHODS: Using data from a large, European-wide observational registry of patients with AF, we implemented different methods to assess clinical complexity, according to a) the presence of at least one determinants among frailty, multimorbidity, polypharmacy; b) levels of complexity, as identified by a k-medoids cluster analyses based on age, frailty, number of comorbidities and number of drugs received; and c) phenotypes of complexity, evaluated through latent class analysis-defined comorbidity patterns, and based on 18 chronic cardiovascular and non-cardiovascular comorbidities/conditions. The association of clinical complexity with major outcomes was evaluated through multiple-adjusted Cox regression analyses at 2 years of follow-up; a composite of all-cause death and major adverse cardiovascular events was the primary outcome for our analyses, with other secondary outcomes also analysed. We additionally evaluated the associations of phenotypes of complexity with use of oral anticoagulant (OAC) at baseline. Finally, we evaluated the association of the adherence to the ABC pathway (either full adherence vs. non-adherence, or adherence to 2-3 vs. 0-1 ABC criteria) on the risk of the primary outcome across different definitions of complexity. RESULTS: 9,966 AF patients were included in the first analysis; of these, 83.2% were considered as clinically complex, presenting with at least one among frailty, multimorbidity, or polypharmacy; cluster analysis identified 44.5% of patients with a high clinical complexity level, while others presented with moderate complexity. At multiple-adjusted Cox-regression analysis, clinically complex patients had a higher risk of the primary outcome of all-cause death and MACE (Hazard Ratio [HR]: 1.76, 95% Confidence Interval [CI]: 1.36-2.28), as well as of other secondary outcomes; similar results were found when comparing patients in the high vs. moderate complexity cluster (HR 1.87, 95%CI: 1.59-2.19 for primary outcome). Adherence to the ABC pathway was associated with significant lower risk of the primary composite outcome in clinically complex patients (HR: 0.70, 95%CI: 0.58-0.85) as well as in the high complexity cluster (HR: 0.76, 95%CI: 0.60-0.96), without significant interaction observed in patients with vs. without clinical complexity (p for interaction [pint1]=0.325) and in high vs. moderate complexity cluster (pint=0.497). Similar results were observed when comparing adherence to 2-3 vs. 0-1 ABC criteria. When we explored phenotypes of complexity, we identified 5 comorbidity patterns, with increasingly complex clinical profiles, as follows: “low-morbidity” pattern (which accounted for 46.1% of patients); “cardiovascular” pattern (25.0%) “metabolic pattern” (11.3%); “heart failure” pattern (9.7%); and a “multisystemic” pattern (8.0%). Phenotypes were heterogeneously associated with use of OAC, and showed an increased risk of major outcomes as the complexity of the clinical profile increase; highest magnitude of risk increase for the primary composite was observed in the heart failure and multisystemic patterns (HR: 2.18, 95%CI: 1.74-2.72 and HR: 2.14, 95%CI: 1.62-2.82, respectively). Adherence to the ABC pathway was associated with similar and consistent reduction in the risk of the primary outcome across phenotypes of complexity (pint=0.885); conversely, adherence to 2-3 vs. 0-1 ABC criteria was associated with highest magnitude of risk reduction in the metabolic and multisystemic patterns (pint=0.011). CONCLUSIONS: Clinical complexity is common in patients with AF and is heterogeneously associated with clinical management and prognosis. Evaluation of clinical complexity through integration of different determinants and comorbidities may offer a promising approach in research and real-world practice. An integrated and holistic approach as the ABC pathway is associated with improved prognosis in clinically complex patients with AF, representing a pragmatic and effective intervention to address the increasingly complex health needs of patients with AF.