Abstract Background It is thought that infrabony defect morphology affects the outcome of periodontal regenerative surgery. However, this has not been systematically investigated. Aims To investigate how well defect morphology is described in papers reporting regenerative therapy of periodontal infrabony defects and to investigate its effect on clinical and radiographic outcomes. Materials and Methods A search was conducted in 3 electronic databases for publications reporting clinical and radiographic outcomes of periodontal intra‐bony defects after regenerative therapy, divided by defect morphology. Results The initial search resulted in 4487 papers, reduced to 143 after first and second screening. Fifteen of these publications were suitable for a fixed‐effects meta‐analysis. Initial defect depth was found to influence radiographic bone gain 12 months post‐surgery, while narrower angles and increased number of walls influenced both radiographic bone gain and clinical attachment level (CAL) gain at 12 months. These associations seemed to occur irrespective of biomaterials used. Risk of bias ranged from low to high. Conclusion Deeper defects with narrower angles and increased number of walls exhibit improved CAL and radiographic bone gain at 12 months post‐regenerative surgery. More data are needed about other aspects of defect morphology such as extension to buccal/lingual surfaces.