Background: Growth factors are generally accepted to be essential mediators of tissue repair via well-established mechanisms of action that include stimulatory effects on angiogenesis and cellullar proliferation, ingrowth, differentiation, and matrix biosynthesis. The aim of this study was to evaluate in a large-scale, prospective, blinded, and randomized controlled clinical trial the safety and effectiveness of purified recombinant human platelet-derived growth factor (rhPDGF-BB) mixed with a synthetic beta-tricalcium phosphate (β-TCP) matrix for the treatment of advanced periodontal osseous defects at 6 months of healing. Methods: Eleven clinical centers enrolled 180 subjects, each requiring surgical treatment of a 4 mm or greater intrabony periodontal defect and meeting all inclusion and exclusion criteria. Subjects were randomized into one of three treatment groups: 1) β-TCP + 0.3 mg/ml rhPDGF-BB in buffer; 2) β-TCP + 1.0 mg/ml rhPDGF-BB in buffer; and 3) β-TCP + buffer (active control). Safety data were assessed by the frequency and severity of adverse events. Effectiveness measurements included clinical attachment levels (CAL) and gingival recession (GR) measured clinically and linear bone growth (LBG) and percent bone fill (% BF) as assessed radiographically by an independent centralized radiology review center. The area under the curve (AUC), an assessment of the rate of healing, was also calculated for CAL measurements. The surgeons, clinical and radiographic evaluators, patients, and study sponsor were all masked with respect to treatment groups. Results: CAL gain was significantly greater at 3 months for group 1 (rhPDGF 0.3 mg/ml) compared to group 3 (β-TCP + buffer) (3.8 versus 3.3 mm; P = 0.032), although by 6 months, this finding was not statistically significant (P = 0.11). This early acceleration of CAL gain led to group 1 exhibiting a significantly greater rate of CAL gain between baseline and 6 months than group 3 as assessed by the AUC (68.4- versus 60.1-mm weeks; P = 0.033). rhPDGF (0.3 mg/ml)-treated sites also had significantly greater linear bone gain (2.6 versus 0.9 mm, respectively; P