Patient Characteristics and Clinical Factors Affecting Seroma Volume

Abstract

Partial breast irradiation (PBI) has increased in utilization recently, with the postoperative seroma and clips used to guide target volumes. The ideal post-surgical timing to perform computed tomography (CT)-based treatment planning for this technique is unclear. Prior studies have examined change in volume over time from surgery, however, the effect of patient characteristics on seroma volume has not been previously reported. We sought to investigate patient and clinical factors that may contribute to larger post-surgical seromas, and therefore would predict for larger PBI volumes.212 consecutive women with invasive or in situ breast cancer underwent planning CT after breast-conserving surgery at a single institution with four radiation centers during the year 2020. 215 total lumpectomy cavities were contoured per institutional standard. Contoured volume was computed using the treatment planning system. Generalized estimating equation methods were used to analyze the combined data, using treatment center as the cluster effect. Univariate and multivariate analyses (MVA) were performed to evaluate the associations between the seroma volume and patient and clinical factors.Distribution among sites were 61 patients (28.4%), 24 patients (11.2%), 57 patients (26.5%), and 73 patients (34.0%). Patient characteristics included median age 61.0 years (range, 30-85) and median BMI 30.1 kg/m² (17.3-53.9). 94% of patients did not receive neoadjuvant hormone therapy, 47.9% did not have hypertension, 93.5% of patients did not have coronary artery disease, 34.4% were treated in prone position, mean interval from surgery to CT simulation was 58.0 days ± 49.1, and mean seroma volume was 43.3 cm3 ± 46.1. Longer interval from surgery was significantly associated with smaller seroma volume, P = 0.0360. Age, BMI, hypertension, position, coronary artery disease, and use of neoadjuvant hormone therapy remained significant on MVA (P < 0.05 for all), and thus were independently associated with seroma volume. Increased age, BMI, and prone position vs. supine were associated with larger mean seroma volume; absence of hypertension, absence of coronary artery disease, and absence of neoadjuvant hormone therapy were associated with decreased seroma volume. Surgical re-excision, oncoplastic reduction, prior breast surgery, receipt of chemotherapy, race, diabetes, smoking status, and presence of clips were not significantly associated.Increased age, higher BMI, hypertension, coronary artery disease, and prone position were associated with larger seroma volume, and longer interval from surgery to simulation was associated with smaller seroma volume. Especially relevant for patients planned to receive PBI, these data may be used to select patients for which longer time to simulation may result in smaller seroma volumes and therefore smaller PBI target volumes. Larger datasets and prospective evaluation would be ideal to confirm these hypotheses.