Afacan 2020.

Study characteristics
Methods	Study design: RCT with 3‐arm parallel design Recruitment period: October 2006–May 2008 Setting: University Dental Hospital, Turkey Number of centres: 1 Funding source: The Scientific and Technological Research Council of Turkey, grant/award number 106S212
Participants	Inclusion criteria: diagnosis of chronic periodontitis with PPD ≥ 6 mm Exclusion criteria: systemic disease or on antibiotics from 6 months before or during study Age: 35–60 years Sex: 24 F (FMD: 8; FMS: 7; control: 9), 36 M (FMD: 12; FMS: 13; control: 11) Smokers: 0 Number randomised: 68 (FMD: 23; FMS: 22; control: 23) Number evaluated: 60 (FMD: 20; FMS: 20; control: 20)
Interventions	Comparison: FMD vs control; FMS vs control; FMD vs FMS FMD group: FMD US instrumentation in 2 sessions within 24 hours, after instrumentation: tongue brushing: CHX 1%, 1 minute; rinse: CHX 0.2%, 2 × 1 minute; spray pharynx: CHX 0.2%; subgingival: CHX 1%, 3 × within 10 minutes, repeated at day 8. Home: rinse CHX 0.2%, 1 minute, 2 × day; spray tonsils: CHX 0.2%, 2 × day, 2 weeks FMS group: (FMUD) US instrumentation in 2 sessions within 24 hours Control group: (QRP) 4 sessions – 1‐weekly intervals, start 1 Q, hand instruments OHI before study start: no Instruments used: hand and US instruments Time per Q: unknown Maintenance: at 1, 3 and 6 months from last instrumentation Retreatment: none Duration of study: 6 months
Outcomes	Primary outcome: change in GCF biomarker levels Secondary outcomes: PPD, CAL (6 sites per tooth), changes of percentage of pockets with initial PD ≥ 5 mm, changes periodontal pathogen levels Teeth: whole‐mouth recordings with manual probe Pocket depth at baseline: sampling sites (GCF) ≥ 6 mm for selected sites Outcome time reported: 3‐ and 6‐month data used. Baseline, 1, 3 and 6 months measured Other outcomes: PI, PBI, number of single‐rooted teeth, tooth loss, any harm after treatment
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Study co‐ordinator randomly allocated participants into 3 groups using a computer‐generated randomisation list.
Allocation concealment (selection bias)	Unclear risk	Unknown; operator and examiner was the same person.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Clinical measurements, GCF, plaque sampling and all treatments were performed by the same trained investigator in a standardised way. Comment: the same person undertook the interventions and the outcome assessments.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants completed study.
Selective reporting (reporting bias)	Low risk	Data reported on all primary and secondary outcomes.
Other bias	Low risk	Baseline balance good for pocket depth. No apparent other biases.