Methods

3‐arm, non‐blinded, parallel‐group, with 21 days duration of treatment and follow‐up

Participants

Location: Turkey, 1 site

Setting of recruitment and treatment: ear, nose and throat department of Haydarpasa Numune education and research hospital, Istanbul

Sample size: 45

Number randomised (and completed): 15 in oral steroids and INCS, 15 in INCS alone, 15 in oral steroids alone (see notes below)

Participant (baseline) characteristics:

• Mean age (SD): 34.7 ± 16.72 years (average for all 3 groups)

• Gender (male/female): 25/20 (across all 3 groups)

• Main diagnosis: volunteers who received a diagnosis of nasal polyposis

• Polyps status: 100% with polyps

• Polyp grade n (%):

‐ Grade 1: oral steroids and INCS: 1 (6.7%); INCS alone: 3 (13.3%)

‐ Grade 2: oral steroids and INCS: 9 (60%); INCS alone: 5 (33.3%)

‐ Grade 3: oral steroids and INCS: 5 (33.3%); INCS alone: 8 (53.3%)

• Previous sinus surgery status: no information

• Previous courses of steroids: no information

• Positive skin prick test: INCS alone: 7; oral steroids and INCS: 6

Inclusion criteria: none stated. No information on polyp grading criteria (other than Rasp).

Exclusion criteria: patients in whom corticosteroid therapy was contraindicated (i.e. those with diabetes mellitus, hypertension, glaucoma, a history of tuberculosis or emotional instability), as well as those who had received corticosteroids within the last month

Interventions

Oral steroids and INCS (n = 15): oral methylprednisolone, 1 mg/kg and reduced progressively over a 21‐day treatment course

INCS alone (n = 15): no oral steroid treatment

Use of additional interventions (common to both treatment arms): budesonide, unclear method of administration except it states 'intranasal', 400 μg/day, 21 days

Outcomes

Primary outcomes: none reported

Secondary outcomes:

• Polyps size: measured by endoscopic appearance and staging according to the Rasp Classification (1 to 4, 1 = least severe) at 21 days

Other outcomes reported by the study:

• Measurement of nitric oxide levels

Funding sources

No information provided

Declarations of interest

"None declared"

Notes

The trial is a 3‐arm trial comparing "oral steroids alone", "INCS alone" and "oral steroids and INCS". The results for the "oral steroids alone" group are not presented here as they are not relevant to this review.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

High risk

Quote: "Patients were allocated in turn to either the first, second or third treatment group, depending on their order of presentation."

Comment: pg 585, col 1, para 4

Allocation concealment (selection bias)

High risk

Comment: as randomisation was completed based on order of presentation, there is a high risk of bias due to allocation concealment

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Comment: the study was not blinded

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Comment: there is no information regarding blinding of outcome assessment

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Comment: no patients dropped out of the study

Selective reporting (reporting bias)

Low risk

Comment: all outcomes are reported

Other bias

Unclear risk

Comment: Rasp classification used for endoscopic polyp staging. Unclear whether this has been validated and no details on stage used.

Baseline characteristics are not well described but the polyp grade seems to be unevenly distributed (although not statistically significant) with more severe patients in the placebo group, which may have affected the results. The small size of the trial makes it difficult to draw conclusions.

Methods

2‐arm, double‐blind, parallel‐group RCT, with 15‐day duration of oral steroid treatment and 30‐day duration of follow‐up

Participants

Location: Turkey, 2 sites

Setting of recruitment and treatment: paediatric allergy and ear, nose, and throat outpatient clinics of 2 university hospitals

Sample size: 48

Number randomised: 24 in oral steroids and antibiotics, 24 in antibiotics alone

Number completed: 22 in intervention, 23 in antibiotics alone

Participant (baseline) characteristics: note: no. analysed not randomised

• Mean age (SD): oral steroids and antibiotics: 8.5 (2.9); antibiotics alone: 8.0 (2.3)

• Gender (M/F): oral steroids and antibiotics: 14/8 antibiotics alone: 15/8

• Main diagnosis: children with chronic rhinosinusitis

• Polyps status: 0% with polyps

• Previous sinus surgery status: no information

• Previous courses of steroids: information not provided

• Atopy: oral steroids and antibiotics: 8 (36%); antibiotics alone: 10 (43%)

Inclusion criteria: children with CRS; CRS diagnosis made on a basis of sinonasal symptoms and signs present for a period of more than 3 months in the presence of abnormalities on coronal sinus computed tomographic (CT) scans. All patients presented with nasal purulence, postnasal purulence or both and 1 or more of the following symptoms: nasal obstruction, cough, halitosis, headache or facial pain/pressure. They had multiple courses (each 10 to 14 days, > 3 courses) of antimicrobial treatment with at least 2 or more of the following broad‐spectrum antibiotics before entry into the study: amoxicillin‐clavulanic acid, second‐generation cephalosporins (mostly cefuroxime) or clarithromycin

Patients with allergic rhinitis were also included if they also showed purulent rhinorrhoea, postnasal purulence or both

Exclusion criteria:

‐ Systemic corticosteroids in the last 2 months before the study

‐ Systemic antibiotics and inhaler or intranasal corticosteroids in the last 4 weeks before the study

‐ Other respiratory tract disorders (cystic fibrosis, ciliary dyskinesia, nasal polyps, large adenoids and asthma), immune deficiency

‐ Systemic disease, gastroesophageal reflux, aspirin sensitivity

‐ Acquired or congenital sinonasal abnormalities, contraindication to corticosteroid use

‐ Patients with pollen‐induced rhinitis if they were seen during the pollen season

Interventions

Intervention (n = 24): oral methylprednisolone tablets, 15 days according to the following schedule (doses were rounded up to the nearest 4 mg):
‐ 1 mg/kg/day (maximum, 40 mg/day) for 10 days
‐ 0.75 mg/kg/day for 2 days, 0.5 mg/kg/day for 2 days
‐ 0.25 mg/kg/day for 1 day

Comparator group (n = 24): placebo tablets for 15 days

Use of additional interventions (common to both treatment arms): antibiotics (oral amoxicillin/clavulanate) was administered at 45/6.4 mg/kg/day (maximum, 2000/285 mg/day) for 30 days

Outcomes

Primary outcomes:

1. Disease severity, assessed by the patients and their parents by using a visual analogue scale (VAS) (range 0 (no symptoms) to 10 (most severe)). The symptoms scored were: purulent nasal discharge, nasal obstruction, postnasal drainage, halitosis, cough, and facial pain or headache. Individual scores were combined to make a rhinosinusitis symptom score (range 0 to 60) at the end of treatment measured at 30 days.

Secondary outcomes:

2. CT scan, scored using the Lund‐Mackay staging system (0 to 24) measured at 30 days

Other outcomes reported by the study:

• Compliance

• Clinical recovery (definition in paper)

• Relapse (definition in paper)

• "Tolerability was evaluated by means of medical history, physical examination, and measurement of adverse events. Hypertension, edema, weight gain, increase in appetite, gastrointestinal disturbances, nervousness, agitation, psychosis, headache, mood swings, delirium, euphoria, moon face, skin atrophy, bruising, hyperpigmentation, muscle weakness, joint pain, and allergic reactions were defined as clinically significant adverse events"

Funding sources

No information provided

Declarations of interest

"The authors have declared that they have no conflicts of interest."

Notes

—

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "… a random allocation chart based on a table of random numbers."

Comment: pg 349, col 2, para 1

Allocation concealment (selection bias)

Low risk

Quote: "Randomization assignments were kept in sealed envelopes"

Comment: pg 349, col 2, para 1

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Quote: "Placebo tablets contained lactose and were of same size and colour as methylprednisolone (16 mg per tablet). Placebo and methylprednisolone tablets were dispensed in identical packets containing a minimum of 20 tablets each."

"The randomization code was kept by the nursing staff in the pediatric allergy department."

Comment: pg 349, col 2, para 1

Although the paper states that the tablets were equivalent with respect to size and colour, no mention was made about the taste of the tablets. As the taste of methylprednisolone is different to lactose, it may have been obvious which was the treatment. 1 patient in the treatment group dropped out due to unpalatability.

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Comment: the study reports that the participants were blinded and so the patient‐reported outcomes are likely to have been blinded. For the CT scan outcome it is noted that the assessor was blind to treatment and sequence.

No mention of whether the analysis was completed blind.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Comment: although there was low drop‐out (3/48 = 6%) there are no clear reasons provided for 2 of the patients' "protocol violation" or why their results were not included in the analysis

Selective reporting (reporting bias)

High risk

Quote: "No clinically significant adverse events were reported. Twenty seven parents (16 in the methylprednisolone group and 11 in the placebo group) reported that their children's appetite and weight increased after treatment. At the end of the treatment, the mean ± SD changes in patients’ weights from baseline were 0.42 ± 0.26 kg in the methylprednisolone group versus 0.27 ± 0.30 kg in the placebo group. The difference was not significant (P = 0.08)"

Comment: although all of the efficacy outcomes are presented in the results section, the methods section classified increase in appetite and weight gain as clinically significant adverse events and the results indicated that there may be a difference between the groups

The summed scores for the individual symptom scores as presented in the paper do not add up to the total symptom score as presented

Other bias

Unclear risk

Comment: no information regarding the validation of visual analogue scales for reporting CRS symptoms. Nothing was made in the results of the finding that the placebo group benefited substantially from placebo treatment.