The present study will examine the comparative effectiveness of two treatment strategies currently used in the treatment of patients with systolic heart failure presenting with pleural effusion. Patients will be randomized to standard medical treatment only or medical treatment and referral to thoracentesis.

Study hypothesis: A strategy of referring patients with heart failure-related pleural effusion to thoracentesis increases number of days alive outside of hospital over the following 90 days.

Purpose and Rationale: Pleural effusion is a common presentation in patients with heart failure, and the condition is related to a poor prognosis and increased mortality. Therapeutic options consist of intensification of diuretic treatment and invasive drainage of the effusion (thoracentesis). Thoracentesis is a common medical procedure and is often performed on patients with heart failure presenting with a pleural effusion, but there is no randomized evidence to guide the use of thoracentesis in heart failure-related pleural effusion. International guidelines provide no recommendations. Some Danish hospitals use thoracentesis frequently, some rarely. Hence, there is true clinical equipoise and a strong need to assess whether thoracentesis benefits patients or not.

Study Hypothesis: A strategy of referring patients with heart failure-related pleural effusion to thoracentesis increases number of days alive outside of hospital over the following 90 days.

Study Setting:126 adult patients admitted with systolic heart failure and pleural effusion documented by either chest x- ray, ultrasound, computed tomography (CT) or magnetic resonance imaging (MRI) will be randomized 1:1 to medical treatment only or medical treatment and referral to thoracentesis. Thoracentesis will be performed according to local practice. Randomization will be stratified according to whether patients are treated with oral anticoagulation.

Crossover/rescue thoracentesis: For patients in the medical treatment arm whose condition deteriorates to the degree that the participant fulfill any of the study exclusion criteria (eg. increased need of oxygen), thoracentesis may be performed immediately. Some patients randomized to medical treatment only may prove to be too diuretic resistant to achieve an adequate effect of medical treatment alone. The recommended waiting period before performing thoracentesis on a patient in the control group is 5 days from randomization.

Patients may be discharged at the discretion of the treating physician. The follow up period is planned to 90 days.

Inclusion Criteria:

• Left ventricular ejection fraction (LVEF) ≤ 45%
• Non- negligible pleural effusion by x-ray, ultrasound, CT or MRI, suspected to be due to heart failure
• Age ≥ 18 years

Exclusion Criteria:

• Clinically indicated diagnostic thoracentesis (e.g. suspected malignant aetiology)
• Suspected pulmonary or pleural infection (pneumonia symptoms with c- reactive protein>100 mg/L or white blood count>11)
• Recent (<3 months) intrathoracic procedure (including heart or lung surgery, transcatheter aortic valve implantation (TAVI) or thoracentesis)
• Contraindications to thoracentesis according to local guidelines (such as spontaneously elevated international normalized ratio (INR) or thrombocytopenia), patients on oral anticoagulation may be randomized.
• Severe aortic stenosis
• Massive pleural effusion (equivalent to >2/3 of lung on a standing chest x-ray)
• Substantially affected hemodynamics (systolic blood pressure < 85mmHg, need of inotropes) or high oxygen demand (>7L/min)
• Estimated glomerular filtration rate (eGFR) <15ml/min/1.73m2 or dialysis treatment
• Planned or expected admission > 10 days for other condition than heart failure
• Inability to give informed consent

De-identified data necessary to reproduce published results will be made available on reasonable requests.