Recently prepectoral breast reconstruction (PBR) has showed acceptable outcomes in the short-term. There are few evidence on long-term results and variables which could influence surgical safety. So far, no specific guidelines or indications have been developed for prepectoral technique and heterogeneous inclusion criteria had been used in previous reports.This study revises a series of 397 patients. We conducted a retrospective comparative analysis of risk factors and outcomes between patients undergoing direct to-implant (DTI) and patients undergoing two-stages expander-assisted (TSE) PBR. Univariate binary logistic regression was performed to investigate the association between the incidence of postoperative and aesthetic complications and several variables.521 breasts were included in the analysis, with an average follow-up of 38 months. 210 patients underwent DTI and 187 TSE PBR. No statistical differences were found between the two populations in term of the characteristics of patients, surgeries and outcomes. Binary logistic regression found no significant association in the TSE group. In the DTI group, a significant association was found between surgical complications and BMI and adjuvant radiotherapy. The association remained significant only for BMI, when investigated with the onset of aesthetic complications.Lower BMI and adjuvant radiotherapy are significantly associated to a higher risk of developing a surgical complication in DTI PBR. Patients at lowest BMI with DTI are prone to develop an aesthetic complication. According to this analysis, we suggest to carefully choose candidates for PBR and propose new selection criteria for subcutaneous techniques.