Terra 2015.

Methods	Single centre RCT evaluating three different doses of silver nitrate for pleurodesis in malignant pleural effusion (Brazil)
Participants	Inclusion criteria: recurrent and symptomatic MPE (with pleural histological or cytological confirmation); previous CXR showing full lung expansion (> 90%) after chest drainage; Karnofsky performance score > 40; written consent Exclusion criteria: trapped lung after pleural catheter insertion; haemorrhagic diathesis (PT < 50% or platelets < 80,000); active pleural or systemic infection; neoplastic infiltration of the skin at the site of pleural catheter insertion; inability to understand QoL questionnaires; contralateral pleurodesis < 30 days before study entry 60 participants randomised
Interventions	All participants were admitted for five days and had baseline assessment. All had a 14 Fr chest drain inserted under USS guidance prior to randomisation. The randomised interventions were given via the chest tube, which was then clamped for one hour. Drain removed on day 5. The silver nitrate was dissolved in 100 ml distilled water, which was passed through a 0.22 micrometer filter to ensure sterility within six hours of instillation Group 1: 30 ml of 0.3% silver nitrate (90 mg) given as a single dose intrapleurally Group 2: 30 ml of 0.5% silver nitrate (150 mg) given as a single dose intrapleurally Group 3: 60 ml of 0.3% silver nitrate (180 mg) given as a single dose intrapleurally
Outcomes	Primary outcome: occurrence of serious or severe adverse event during follow up Secondary outcomes: systemic inflammation (measured using CRP); chest pain (measured using VAS score); effusion recurrence (defined as need for additional pleural procedures during trial follow up); residual pleural cavity volume (calculated using difference between day 5 and day 30 on CT)
Notes	People with trapped lung excluded from study entry Pleurodesis failure defined as need for additional pleural procedure during follow up Not included in network meta‐analysis
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Block randomisation
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Pharmacy employees and clinicians who instilled the sclerosant were aware of treatment allocation, but these clinicians were not involved in patient follow up. Participants, investigators that followed participants up and rated their complications were blinded to group allocation
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Pharmacy employees and clinicians who instilled the sclerosant were aware of treatment allocation, but these clinicians were not involved in patient follow up. Participants, investigators that followed participants up and rated their complications were blinded to group allocation
Incomplete outcome data (attrition bias) All outcomes	Low risk	LTFU well balanced and justified
Selective reporting (reporting bias)	Low risk	No data provided for MRC dyspnoea score. Otherwise all predefined outcome measures reported
Other bias	Low risk	No other biases identified