Identification and control of significant factors determining clinical outcomes is of paramount importance to improve expected results of a variety of therapeutic procedures. The aim of this investigation was to identify, with a multivariate approach, factors associated with healing outcomes of 3 periodontal surgical procedures in deep intrabony defects. 45 patients with evidence of deep intrabony defects were randomly assigned to 3 treatment groups: access flap (group C), conventional guided tissue regeneration (GTR) with non-resorbable expanded polytetrafluoroethilene (ePTFE) membranes (group B), and GTR with self supporting membranes combined with the modified papilla preservation technique (group A). In both GTR procedures, membranes were positioned coronal to the interproxymal alveolar crest. Primary outcome variables (i.e., probing attachment level gains at 1 year and the amount of newly formed tissue present at membrane removal) were explained in terms of a series of patient, defect morphology and surgical factors, using a multivariate approach. Highly significant treatment effects were observed, indicating that the 3 tested therapeutic modalities resulted in significant differences in primary outcome variables. Detailed analysis assessing the significance of the tested factors in determining the healing outcomes following each procedure was performed with a stepwise elimination approach of non-significant factors. The results indicated that: (i) the need to create and maintain space should be a key objective of regenerative approaches based upon the principles of guided tissue regeneration; (ii) control of patient's oral hygiene and residual periodontal infection in the oral cavity are strongly associated with clinical outcomes of both regenerative and conventional surgical procedures and should receive proper attention.