This is a split-mouth, double-blind randomized controlled trial. Computerized linear radiographic measurements were used to detect infrabony defects treated with open flap debridement (OFD) or OFD associated with enamel matrix derivative (EMD), after 24 months. Ten patients presenting 2 or more defects were selected (43 defects). An individualized film holder was used to take standardized radiographs at baseline and after 24 months. Images were digitized and used to measure the distances from the cemento-enamel junction (CEJ) to the alveolar crest (AC), CEJ to the bottom of the defect (BD) and infrabony defect angle.

This was a split-mouth randomized controlled trial. Patients were recruited among the ones seeking for periodontal treatment at the Post-graduate Clinic of Periodontics using the following inclusion criteria: (1) diagnosis of chronic periodontitis15; (2) presence of at least one pair of interproximal infrabony defects (2-3 walls) adjacent to vital anterior or premolar teeth; (3) absence of 2 and 3 mobility degrees16; (4) probing pocket depth (PPD)≥ 5mm; (5) full-mouth plaque score ≤ 20%17; and (6) keratinized tissue width of at least 2mm. The exclusion criteria were: (1) presence of any systemic disease that could interfere with periodontal treatment; (2) infrabony defects with trans-surgical depth ≤4mm; (3) antibiotic treatment administered during the last 6 months.

The patients participating in the study were volunteers who received and gave informed consent and were included in the study from June to October 2002.

Following initial examination, all patients underwent oral hygiene instruction and full-mouth supra- and sub-gingival scaling and root planning under local anesthesia. Patients were re-evaluated after completion of the initial therapy to determine their response to therapy and to confirm the need for periodontal surgery. Before surgery, for each pair of defects, one defect was randomly assigned to test (EMD) and one to control (OFD) treatment by toss of a coin. Both defects were treated at the same surgical time. When a patient presented with more than one pair of defects, only one pair was treated per day. Following local anesthesia, all sites were treated with reflection of a full thickness mucoperiosteal flap after intra-sulcular incisions. The exposed roots and osseous defects were debrided with hand instruments, and the surgical wound was rinsed with saline.

After that, the flaps of the OFD sites were repositioned and sutured using 5-0 nylon sutures. The EMD sites were dried with non-woven gauze, roots were conditioned with 24% ethylenediaminotetracetic acid (EDTA) gel (pH 6.7) for 2 minutes. The defect was thoroughly rinsed with saline, and EMD gel was applied to the root surfaces according to the manufacturer's instructions. The flaps were then replaced for primary closure and sutured with 5-0 nylon sutures. The sutures were removed after 7 days. All surgeries were performed by the same investigator. All patients were prescribed 0.12% chlorhexidine digluconate and instructed to rinse gently twice a day for 4 weeks. Analgesics were prescribed to be taken as needed, and all patients were seen once a week, for 8 weeks, for professional tooth cleaning. Subsequently, the patients were maintained in a supportive periodontal program (ie, professional tooth cleaning and reinforcement of self-administered oral hygiene measures) at 2-month intervals up to 6-month time and then every 3 months until final examination at 24 months.

Standardized periapical radiographs were taken at baseline evaluation, immediately before surgery and at 24 months follow-up. Individually customized bite blocks employing a reference occlusal stent and film holders were used to obtain reproducible exposed films at each radiographic control. All radiographs were evaluated by a single calibrated examiner, blind to time and treatment.

Analyses of the radiographic outcomes were performed using computerized linear measurements with image analysis software. The radiographs were previously scanned in digital format by a scanner at a resolution of 500dpi/8bits.

The radiographic analysis was based in anatomical landmarks (CEJ, BD and AC) that were identified on the scanned radiographs. All linear measurements were recorded by a blinded, calibrated examiner. The following outcomes were measured at radiographs taken at baseline and after 24 months:

1. Distance from the CEJ to the bottom of the defect (BD). The most coronal area where the periodontal ligament maintained an even width was identified to measure the most apical extension of the infrabony defect;
2. Distance from the CEJ to the bone crest (BC);
3. Infrabony defect angle was defined by two lines that represented the root surface of the involved tooth (CEJ-BD) and the bone defect surface (BD-BC).

The distance from the CEJ to the bottom of the defect (BD) was considered the primary outcome. CEJ-BC and defect angle were secondary outcomes

Inclusion Criteria:

1. diagnosis of chronic periodontitis
2. presence of at least one pair of interproximal infrabony defects (2-3 walls) adjacent to vital anterior or premolar teeth
3. absence of 2 and 3 mobility degrees
4. probing pocket depth (PPD)≥ 5mm
5. full-mouth plaque score ≤ 20%
6. keratinized tissue width of at least 2mm

Exclusion Criteria:

1. presence of any systemic disease that could interfere with periodontal treatment
2. infrabony defects with trans-surgical depth ≤4mm
3. antibiotic treatment administered during the last 6 months -