Hansen 2010.

Methods	2‐arm, double‐blind, placebo‐controlled, parallel‐group RCT with 12 weeks of treatment and 14 weeks of follow‐up
Participants	Location: Netherlands, single site Setting of recruitment and treatment: ENT clinic in the Netherlands (Academic Medical Centre, Amsterdam) Sample size: Number randomised: 10 in intervention, 10 in comparison Number completed: 9 in intervention, 7 in comparison Participant (baseline) characteristics: Age, mean (range): FP: 49.2 (25 to 61); PL: 46.7 (37 to 62) Gender, M/F: FP: 6/4; PL: 8/2 Main diagnosis: recalcitrant CRS without nasal polyps or only cobble‐stoned mucosa Polyps status: no nasal polyps Previous sinus surgery status: all had surgery before Sinus surgery, median (range): FP: 4 (1 to 10); PL: 3 (1 to 8) Polypectomy (%): FP: 2 (20); PL: 0 Previous courses of steroids: no information Other important effect modifiers: Current asthma, n (%): FP: 4 (40); PL: 3 (30) Allergy, n (%): FP: 5 (50); PL: 5 (50) ASA intolerance, n (%): FP: 3 (30); PL: 1 (10) Inclusion criteria: Between 18 and 65 years of age Chronic rhinosinusitis defined as at least a 12‐week history of 2 or more of: blockage/congestion, discharge: anterior/post nasal drip, facial pain/pressure, reduction or loss of sense of smell and either mucopurulent discharge from the middle meatus or oedema/mucosal obstruction primarily in the middle meatus Exclusion criteria: Visible nasal polyps on endoscopy, except cobble‐stoned mucosa Surgical treatment for nasal polyps during the previous 3 months A diagnosis of cystic fibrosis Depot or oral steroids during the previous 2 months A requirement for more than 1000 μg beclomethasone (or equivalent) per day for the treatment of asthma, or not on a stable dose for ≥ 3 months
Interventions	Intervention (n = 10): fluticasone propionate, administered using a breath actuated inhaler (Optinose) 400 µg twice daily (800 µg total daily dose), duration of treatment Comparator group (n = 10): matching placebo, administered twice daily Use of additional interventions (common to both treatment arms): Participants using saline rinses were permitted to continue to do so; 5 in the placebo group and 7 in the treatment group continued using nasal saline irrigation twice daily during the study Loratadine 10 mg tablets provided as rescue medication If a participant experienced a severe acute nasal blockage the investigator could authorise the use of a short course of oxymetazoline drops or spray for a maximum of 7 consecutive days and a maximum total of 10 days during the treatment period. Oxymetazoline was not to be used within 24 hours of a scheduled study visit.
Outcomes	Outcomes of interest in the review: Primary outcomes: 1. Health‐related quality of life, disease‐specific using RSOM‐31. A reduction in the average total symptom impact score > 1 is considered clinically relevant. 2. Disease severity symptom score, measured using a. A 10 cm VAS (not troublesome to most troublesome imaginable) "How troublesome are your symptoms of rhinosinusitis?" b. A diary (0 to 3 scale): 0 (none), 1 (mild – symptoms present but not troublesome), 2 (moderate – symptoms frequently troublesome but not interfering with daily activity or night‐time sleep) or 3 (symptoms troublesome and interfering with daily activity or night‐time sleep) to record nasal blockage, nasal discomfort and rhinitis symptoms. Participants also recorded sense of smell: 0 (normal), 1 (slightly impaired), 2 (moderately impaired) or 3 (absent). 3. Significant adverse effect: epistaxis Secondary outcomes: 4. Endoscopy (polyps size or overall score) using the Lund‐Mackay score Other outcomes reported by the study: Peak nasal inspiratory flow (PNIF) Acoustic rhinometry MRI scans of the paranasal sinuses Use of rescue medication
Notes	Study had a 14‐ to 16‐day treatment‐free run‐in period at the beginning
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "…subjects who met the eligibility criteria were randomized 1:1 …" Comment: no information on randomisation method provided
Allocation concealment (selection bias)	Unclear risk	Comment: no information on how to maintain allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "…The Opt‐FP and placebo devices were identical in appearance. The spray pump in the Opt‐FP contained … Placebo matched FP exactly, except for the active ingredient... To deliver a dose of FP 400 μg bd. or placebo, subjects made two administrations per nostril, morning and evening. " Comment: matched placebo – identical‐looking and same formulation except for not including the active ingredient. Patients using the placebo administered it exactly the same number of times as the treatment group.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: as above Comment: key outcomes are patient‐reported – should remain well blinded until end of study
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "…The majority of subjects (9 Opt‐FP, 7 PBO) completed the study." Comment: the drop‐out rate is high (20%) and unbalanced (10% in active group and 30% in placebo group) enough to affect the findings of this small study (results were presented as medians). All withdrawals were related to adverse effects or worsening of symptoms.
Selective reporting (reporting bias)	Unclear risk	Comment: unclear why study reported "combined nasal symptoms" (rhinitis and blockage) only, whereas other symptoms (discomfort and smell) were reported separately. This was not prespecified. Reporting of some outcomes was confusing or incomplete. The results for the for "symptoms of rhinosinusitis" as measured on a 10 cm VAS were reported as having reduced 13 points in the treatment group. It was not clear if this was a mistake or whether it was 13%. For endoscopic evaluation, the study only showed data for the oedema score, which was statically significant, but not the nasal discharge score, which was not statistically significant.
Other bias	Low risk	Comment: instruments for the main patient‐reported symptoms were validated