Maskell 2004.

Methods	Single centre RCT comparing pleurodesis using mixed particle Talc (>50% of particles are <20μm) vs graded Talc (<50% of particles are <20μm) (UK)
Participants	Inclusion: Symptomatic pleural effusion, proven to be malignant by cytology or pleural biopsy (all cell types). Exclusion: Expected survival <6 weeks; bleeding diathesis contraindicating intercostal drain insertion; extensive trapped lung; previous ipsilateral pleurodesis; Age <18; Inability to give informed consent. 48 patients randomised.
Interventions	12Fr intercostal drain inserted. Drainage until <150ml/day output. Agent instilled and left in for 2 hours, before suction being applied. Drain removed after 48 hours. Mixed particle talc group: >50% of talc particles are <20μm. Single 4g intrapleural dose. Graded talc group: <50% of talc particles are <20μm. Single 4g intrapleural dose.
Outcomes	Change in Aa gradient 48hours post pleurodesis breathing air Change in PaO2 at 48hours post pleurodesis Clinical efficacy of pleurodesis at 3 months Presence/absence of fever at 48hours Change in CRP Change in IL8
Notes	Patients with trapped lung excluded. Pleurodesis success defined as no re‐accumulation of pleural fluid sufficient to require drainage. Paper presented 2 trials and only trial 2 was relevant to this review (trial 1 was RCT of mixed talc vs tetracycline, but pleurodesis success data was not collected) Not included in network meta‐analysis
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	pre‐sealed numbered, opaque, sealed envelopes with stratification
Allocation concealment (selection bias)	Low risk	pre‐sealed numbered, opaque, sealed envelopes with stratification
Blinding of participants and personnel (performance bias) All outcomes	Low risk	'investigators and patients blind to treatment allocation' (personal communication with authors)
Blinding of outcome assessment (detection bias) All outcomes	Low risk	'investigators and patients blind to treatment allocation' (personal communication with authors)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing data justified and balanced between the 2 groups (3 patients LTFU)
Selective reporting (reporting bias)	Unclear risk	The study comprised of two sections and pleurodesis success only reported for the particle size section. The RCT of talc/tetracycline did not report pleurodesis success but this was not one of the pre‐defined outcome measures.
Other bias	Low risk	No other biases identified