Tonetti 2002.

Methods	1‐year follow‐up parallel‐group study including 2 groups with 172 patients in total. 6 drop outs at 1 year. 3 patients withdrew consent before surgery. 3 patients (2 from the test and 1 from the control group) were unable to comply with the follow up for reasons independent from the treatments.
Participants	Patients in good general health and motivated for good oral hygiene. Teeth with IBD greater or equal to 3 mm in presence of 2 to 3 mm of keratinized gingiva on the buccal aspect. Heavy smokers (> 20 cigarettes per day) were excluded. 1‐, 2‐ and 3‐wall defects were included. Age ranging between 39 and 57; females were 54.2% in the test and 60.2% in the control groups. Patients were recruited both from university dental clinics and private practices.
Interventions	Emdogain versus flap surgery.
Outcomes	FMPS and FMBS. For experimental teeth only: PAL, PPD, REC at baseline and 1 year. Tooth loss and postoperative infections. Additional intrasurgical measurements were taken. 1‐year data used. Postoperative morbidity, patient satisfaction, aesthetics and several other patient‐centred outcomes were evaluated.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "All subjects were assigned a patient number, and were randomly assigned to one of the two treatment regiments. Assignment was performed by a central randomization facility using a custom‐made program based on balanced random permuted blocks".
Allocation concealment (selection bias)	Low risk	Author informed us that the allocation to the intervention groups was concealed. During surgery, after debridement, a sealed opaque envelope containing the randomisation code was opened.
Blinding (performance bias and detection bias) All outcomes	High risk	Quote: "In each center, the examiner and the therapist were identical".
Incomplete outcome data (attrition bias) All outcomes	Low risk	All outcome data are presented in Table 2: Clinical outcomes at 1 year. Comment: No missing data. Drop outs are explained adequately.
Selective reporting (reporting bias)	Low risk	All the pre‐specified outcomes are reported in Table 2: Clinical outcomes at 1 year. No teeth were extracted. No infectious complications were observed. The aesthetics evaluation was reported in Tonetti et al 2004 JCP 31:1092‐8. Changes in bone level were not assessed.
Other bias	Low risk	The trial was partially supported with a research grant from the manufacturer. We do not think that this has affected the outcome of the trial.