Abstract Background Immediate implant-based breast reconstruction (IBBR) following nipple-sparing (NSM) or skin-sparing mastectomy (SSM) increasingly requires individualized technique selection. Choosing among prepectoral direct-to-implant (DTI), submuscular DTI, and two-stage expander reconstruction remains complex. This study analyses outcomes from a large single-centre cohort to define a pragmatic, evidence-based decision-making algorithm. Methods A retrospective monocentric review included all consecutive IBBR procedures performed between 2018 and 2024. Patients were allocated to three groups: prepectoral DTI (n = 601), submuscular DTI (n = 142), and two-stage expander reconstruction (n = 44). Demographic, oncologic, and surgical variables were collected. Major complications were evaluated using unadjusted analyses and inverse-probability-of-treatment weighting. A dedicated subgroup analysis was performed for patients receiving postmastectomy radiotherapy (PMRT). Results A total of 1,093 breasts were reconstructed. Weighted analysis showed similar overall major-complication rates for prepectoral and submuscular DTI (7.3% vs 8.4%, p = 0.278), while expanders demonstrated the lowest adjusted risk (0.7%). Within the PMRT subgroup, prepectoral DTI did not show increased infection, dehiscence, or seroma compared with the other techniques; infection was significantly lower than in submuscular DTI (2.7% vs 14.3%, p = 0.020). Capsular contracture was markedly less frequent in prepectoral DTI and remained significantly reduced after multivariable adjustment (OR 1 reference; OR 9.03 for submuscular DTI; OR 3.25 for expanders). Conclusions Our resulting algorithm offers a structured, patient-specific approach to optimise outcomes in immediate implant-based breast reconstruction. Level of Evidence IV This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .
Taraschi et al. (Tue,) studied this question.