Plastic Surgery

EHR-integrated AI risk scores did not increase patient conversations about graft-loss options

In a single-center randomized trial of 76 kidney transplant recipients, passive EHR display of a machine-learning 1-year graft-loss risk did not change patient-reported conversations (39% vs 40%, p=1.00). Secondary outcomes including shared decision-making metrics, clinician–patient relationship measures, and distress showed no significant differences. User feedback noted low and variable tool uptake and workflow barriers as likely reasons for null effects. Implication: passive availability of risk estimates alone is insufficient; interventions must actively integrate into workflows and explicitly support SDM. 1

Surgical root coverage improves professional and patient esthetic outcomes for multiple recessions

Meta-analysis of 32 RCTs (1012 patients, 3589 recessions) found mean Root Coverage Esthetic Score 8.31 and patient VAS satisfaction 8.59. Mean root coverage averaged 82.6% and complete root coverage pooled at 62.7% across techniques. Postoperative pain was low (VAS mean 2.67), and gains in attachment and keratinized tissue were consistent. Applicability: CAF and tunnel approaches with connective tissue grafts or substitutes reliably improve esthetics; counsel patients about likelihood of complete coverage. 2

Reduced preoperative radiotherapy (36 Gy) gave excellent 5-year local control in myxoid liposarcoma

The phase 2 DOREMY trial (n=90, median follow-up 66 months) reported 5-year local recurrence-free survival 97.4%. Five-year progression-free, disease-specific, and overall survival were 81.0%, 89.5%, and 88.5%, respectively. Wound complications occurred in 21% (16% required intervention), and late grade 2–3 toxic effects were uncommon. Caveat: single-group, nonrandomized design—discuss risks and alternatives with patients in multidisciplinary teams. 3