Symptoms and signs in patients with heart failure: Association with 3-month hospitalisation and mortality

Objectives: To determine the association between symptoms and signs reported in primary care consultations following a new diagnosis of heart failure (HF), and 3-month hospitalisation and mortality. Design: Nested case-control study with density-based sampling. Setting: Clinical Practice Research Datalink, linked to hospitalisation and mortality (1998-2020). Participants: Database cohort of 86 882 patients with a new HF diagnosis. In two separate analyses for (1) first hospitalisation and (2) death, we compared the 3-month history of symptoms and signs in cases (patients with HF with the event), with their respective controls (patients with HF without the respective event, matched on diagnosis date (& plusmn;1 month) and follow-up time). Controls could be included more than once and later become a case. Main outcome measures: All-cause, HF and non-cardiovascular disease (non-CVD) hospitalisation and mortality. Results: During a median follow-up of 3.22 years (IQR: 0.59-8.18), 56 677 (65%) experienced first hospitalisation and 48 146 (55%) died. These cases were matched to 356 714 and 316 810 HF controls, respectively. For HF hospitalisation, the strongest adjusted associations were for symptoms and signs of fluid overload: pulmonary oedema (adjusted OR 3.08; 95% CI 2.52, 3.64), shortness of breath (2.94; 2.77, 3.11) and peripheral oedema (2.16; 2.00, 2.32). Generic symptoms also showed significant associations: depression (1.50; 1.18, 1.82), anxiety (1.35; 1.06, 1.64) and pain (1.19; 1.10, 1.28). Non-CVD hospitalisation had the strongest associations with chest pain (2.93; 2.77, 3.09), fatigue (1.87; 1.73, 2.01), general pain (1.87; 1.81, 1.93) and depression (1.59; 1.44, 1.74). Conclusions: In the primary care HF population, routinely recorded cardiac and non-specific symptoms showed differential risk associations with hospitalisation and mortality. & copy; Author(s) (or their employer(s)) 2023. No commercial re-use. See rights and permissions. Published by BMJ.