Middleton 2015.

Study characteristics
Methods	Randomised, double‐blind, placebo‐controlled, pilot study. Parallel design. Duration: 1 month. Location: Australia.
Participants	Inclusion criteria: young people hospitalised with a pulmonary exacerbation (defined by Fuchs criteria). Exclusion criteria: FEV1 on admission < 40%, oxygen requirement, oral corticosteroids, recent commencement of mucolytic agents and failure to tolerate a test dose of the randomised study drug. 23 participants (8 ‐ 18 years) recruited. Both groups had 11 participants. Mean (SD) age: mannitol group = 13.2 (3.7) years; control group = 14.4 (3.0) years. Gender split: mannitol group = 45% male, 55% female; control group = 36% male; 64% female. Control group had lower mean (SD) admission FEV1 (58.0 (13.0)% versus 73.1 (15.6)% predicted, P < 0.05) and mean (range) FEF25-75 (25.7 (20.1 – 71.8)% versus 44.1 (22.8 – 92.43)% predicted, P < 0.05) than the IDPM group. All other parameters were equivalent on admission.
Interventions	Treatment: inhaled dry‐powder mannitol (10x 40 mg capsules) 2x daily. Control: very low‐dose inhaled dry‐powder mannitol (10x 5 mg non‐respirable mannitol) 2x daily. Administered for 12 consecutive days prior to airway clearance treatments supervised by a physiotherapist. Used as an adjunct to intravenous antibiotics.
Outcomes	As this was a pilot study, primary and secondary outcomes were not specified, outcomes measured were: clinical status (CFCS and CFQ‐R); lung function and cardiopulmonary exercise parameters (peak oxygen consumption, ventilation efficacy, carbon dioxide production). Outcome parameters were assessed at day 7, day 14 (discharge from hospital) and 1 month (follow‐up).
Funding Source	Funded by Pharmaxis Australia but all aspects of the research study, design and analysis were conducted without input from Pharmaxis.
Conflicts of Interest	The study authors declare no conflicts of interest.
Notes	Additional unpublished information and results provided from conference presentation slides from the authors. A study‐drug tolerance test was conducted and deemed acceptable if baseline FEV1 decreased by < 50% immediately or < 20% 15 min after the test dose, or if oxygen saturation remained > 89%. Inhaled dry‐powder mannitol was tolerated well with 22 participants completing the study; 1 participant was excluded due to vomiting after the test dose.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Blinded randomisation was performed (1:1) ratio via random number generation without stratification.
Allocation concealment (selection bias)	Unclear risk	Randomisation was described as 'blinded' but not discussed how this was achieved.
Blinding (performance bias and detection bias) Participants	Unclear risk	The study was described as 'double‐blind' but not discussed exactly who was blinded and how the blind was achieved.
Blinding (performance bias and detection bias) Clinicians	Unclear risk	The study was described as 'double‐blind' but not discussed exactly who was blinded and how the blind was achieved.
Blinding (performance bias and detection bias) Outcome assessors	Unclear risk	The study was described as 'double‐blind' but not discussed exactly who was blinded and how the blind was achieved.
Incomplete outcome data (attrition bias) All outcomes	Low risk	22 out of 23 participants completed the study and contributed data to analysis. 1 participant failed the tolerance test and was excluded.
Selective reporting (reporting bias)	Low risk	All important outcomes of the pilot study seem to be reported.
Other bias	High risk	The authors note a relatively small sample size which prevents 'strong conclusions about clinical utility in this setting. The study population included only those with CF who passed the tolerance test and not all potential participants with CF.