Patient-reported outcomes and their predictors at 2- and 3-year follow-up after immediate latissimus dorsi breast reconstruction and adjuvant treatment.

Background The aim of this study was to estimate the impact 2 and 3 years after surgery of implant-assisted latissimus dorsi (LDI) and autologous latissimus dorsi (ALD) flap breast reconstructions on patient-reported outcomes (PROs), and, secondarily, to determine whether baseline characteristics can predict PROs. Methods This was a multicentre prospective cohort study. The European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Core Questionnaire (QLQ-C30) and breast cancer module (QLQ-BR23), Functional Assessment of Cancer Therapy – Breast (FACT-B), and Hospital Anxiety and Depression Scale (HADS) PROs were completed before surgery and at 2 and 3 years after breast reconstruction. The effects of LDI and ALD, adjusted for baseline clinicodemographic characteristics, were estimated with multiple linear regressions. Effect sizes above 0·5 were considered clinically important. Results Some 206 patients (LDI 93, ALD 113) were recruited in 2007–2013; 66·5 per cent were node-negative and 34·6 per cent received radiotherapy. Women with adverse clinicopathological factors were more likely to have received radiotherapy and to undergo ALD. Patients in both surgical groups showed clinically important effects at 2 and 3 years, including improvements in emotional scales, but worse physical functioning, social well-being, body image and anxiety. Radiotherapy adversely affected social functioning at 2 years (P = 0·002). Women undergoing ALD reconstruction had significantly improved sexual functioning at 3 years (P = 0·003) relative to those who had LDI procedures, even after adjusting for case mix (P = 0·007). At 3 years, younger women experienced worse physical well-being than older women (P = 0·006), and chemotherapy was associated with worse arm symptoms (P = 0·005). Conclusion Clinically important changes occurred in physical functioning, breast symptoms, body image and psychological distress. These results will guide selections of key PRO domains and sample-size calculation of future studies.