Existing Psychiatric Diagnoses Among Breast Cancer Patients Interact with Outcomes After Autologous and Implant-Based Bilateral Breast Reconstruction: A Propensity Score Matched Analysis

Background

Breast reconstruction is an integral postoncologic procedure that has been associated with improved mental health and psychological outcomes. The possible interaction between existing psychiatric diagnoses hospital courses and postoperative complications warrants further exploration.

Methods

Bilateral breast reconstruction patients were identified from the 2016 to 2018 Healthcare Cost and Utilization Project-National Inpatient Sample (HCUP - NIS). Number and type of psychiatric diagnoses within the cohort were then evaluated using a host of ICD-10 codes. A propensity score analysis was applied to control for confounding variables such as demographics, existing comorbidities, and hospital characteristics. A binary logistic regression model was then used to identify the prediction value of psychiatric diagnosis and its interaction with modality of reconstruction for objective outcomes like length of hospital stay, treatment charge, and postoperative complications.

Results

A total of 10,114 patients were identified as the final cohort of breast reconstruction patients. 2621 (25.9%) patients possessed an average of 1.4 ± 0.6 existing psychiatric diagnoses. Presence of at least 1 psychiatric diagnosis was a strong predictor alone for extended length of stay (OR: 1.34, 95% CI: 1.28-1.41, P < .001) and occurrence of postoperative complications (OR: 1.31, 95% CI: 1.21-1.41, P < .001). Psychiatric diagnosis displayed a significant interaction with modality of breast reconstruction and conferred a lower increase in risk of extended length of stay in autologous reconstruction when compared to implant-based reconstruction (OR: 0.80, 95% CI: 0.72-0.89, P < .001).

Conclusion

Existing psychiatric diagnoses were shown to strongly predict and modulate risk of adverse postoperative outcomes depending on modality of reconstruction.