METHODS: AMPLE-2 was an open-label randomised trial involving 11 centres in Australia, New Zealand, Hong Kong, and Malaysia. Patients with symptomatic malignant pleural effusions were randomly assigned (1:1) to the aggressive (daily) or symptom-guided drainage groups for 60 days and minimised by cancer type (mesothelioma vs others), performance status (Eastern Cooperative Oncology Group [ECOG] score 0-1 vs ≥2), presence of trapped lung, and prior pleurodesis. Patients were followed up for 6 months. The primary outcome was mean daily breathlessness score, measured by use of a 100 mm visual analogue scale during the first 60 days. Secondary outcomes included rates of spontaneous pleurodesis and self-reported quality-of-life measures. Results were analysed by an intention-to-treat approach. This trial is registered with the Australian New Zealand Clinical Trials Registry, number ACTRN12615000963527.

FINDINGS: Between July 20, 2015, and Jan 26, 2017, 87 patients were recruited and randomly assigned to the aggressive (n=43) or symptom-guided (n=44) drainage groups. The mean daily breathlessness scores did not differ significantly between the aggressive and symptom-guided drainage groups (geometric means 13·1 mm [95% CI 9·8-17·4] vs 17·3 mm [13·0-22·0]; ratio of geometric means 1·32 [95% CI 0·88-1·97]; p=0·18). More patients in the aggressive group developed spontaneous pleurodesis than in the symptom-guided group in the first 60 days (16 [37·2%] of 43 vs five [11·4%] of 44, p=0·0049) and at 6 months (19 [44·2%] vs seven [15·9%], p=0·004; hazard ratio 3·287 [95% CI 1·396-7·740]; p=0·0065). Patient-reported quality-of-life measures, assessed with EuroQoL-5 Dimensions-5 Levels (EQ-5D-5L), were better in the aggressive group than in the symptom-guided group (estimated means 0·713 [95% CI 0·647-0·779] vs 0·601 [0·536-0·667]). The estimated difference in means was 0·112 (95% CI 0·0198-0·204; p=0·0174). Pain scores, total days spent in hospital, and mortality did not differ significantly between groups. Serious adverse events occurred in 11 (25·6%) of 43 patients in the aggressive drainage group and in 12 (27·3%) of 44 patients in the symptom-guided drainage group, including 11 episodes of pleural infection in nine patients (five in the aggressive group and six in the symptom-guided drainage group).

INTERPRETATION: We found no differences between the aggressive (daily) and the symptom-guided drainage regimens for indwelling pleural catheters in providing breathlessness control. These data indicate that daily indwelling pleural catheter drainage is more effective in promoting spontaneous pleurodesis and might improve quality of life.