Gerson M Struik1, Nienke Hoekstra2, Taco M Klem3, Ali Ghandi4, Gerda M Verduijn2, Annemarie T Swaak-Kragten2, Alja Schoonbeek2, Kim C de Vries2, Margriet A Sattler2, Kees Verhoef5, Erwin Birnie6, Jean-Philippe Pignol7. 1. Franciscus Gasthuis and Vlietland, Department of Surgery, Rotterdam, the Netherlands; Erasmus MC, Department of Radiation Oncology, Rotterdam, the Netherlands. 2. Erasmus MC, Department of Radiation Oncology, Rotterdam, the Netherlands. 3. Franciscus Gasthuis and Vlietland, Department of Surgery, Rotterdam, the Netherlands. 4. Franciscus Gasthuis and Vlietland, Department of Radiology, Rotterdam, the Netherlands. 5. Erasmus MC, Department of Surgery, Rotterdam, the Netherlands. 6. Franciscus Gasthuis and Vlietland, Department of Statistics and Education, Rotterdam, the Netherlands. 7. Erasmus MC, Department of Radiation Oncology, Rotterdam, the Netherlands; Dalhousie University, Department of Radiation Oncology, Halifax, Canada. Electronic address: jean-philippe.pignol@dal.ca.
Abstract
BACKGROUND AND PURPOSE: During oncoplastic breast-conserving surgery (BCS), the surgical cavity is closed to reduce seroma formation. This makes the radiotherapy target definition using clips challenging, leading to poor inter-observer agreement and potentially geographical misses. We hypothesize that injecting a radiopaque hydrogel in the lumpectomy cavity before closure improves radiotherapy target definition and agreement between observers. MATERIALS AND METHODS: Women undergoing BCS in a single university hospital were prospectively accrued in the study. Three to 9 ml of iodined PolyEthylene Glycol (PEG) hydrogel and clips were inserted in the lumpectomy cavity. A CT-scan was performed at 4 to 6 weeks. CT images of BCS patients with standard clips only were used as control group, matched on age, specimen weight, and distance between clips. Six radiation oncologists delineated the tumor bed volumes and rated the cavity visualization scores (CVS). The primary endpoint was the agreement between observers measured using a Conformity Index (Cx). RESULTS: Forty-two patients were included, 21 hydrogel procedures and 21 controls, resulting in 315 observer pairs. The feasibility of the intervention was 100%. The median Cx was higher in the intervention group (Cx = 0.70, IQR [0.54-0.79]) than in the control group (Cx = 0.54, IQR [0.42-0.66]), p < 0.00, as were the CVS (3.5 [2.5-4.5] versus 2.5 [2-3.5], p < 0.001). The rate of surgical site infections was similar to literature. CONCLUSIONS: The use of radiopaque PEG enables to identify the lumpectomy cavity, resulting in a high inter-observer agreement for radiotherapy target definition. This intervention is easy to perform and blend well into current practice.
BACKGROUND AND PURPOSE: During oncoplastic breast-conserving surgery (BCS), the surgical cavity is closed to reduce seroma formation. This makes the radiotherapy target definition using clips challenging, leading to poor inter-observer agreement and potentially geographical misses. We hypothesize that injecting a radiopaque hydrogel in the lumpectomy cavity before closure improves radiotherapy target definition and agreement between observers. MATERIALS AND METHODS:Women undergoing BCS in a single university hospital were prospectively accrued in the study. Three to 9 ml of iodined PolyEthylene Glycol (PEG) hydrogel and clips were inserted in the lumpectomy cavity. A CT-scan was performed at 4 to 6 weeks. CT images of BCS patients with standard clips only were used as control group, matched on age, specimen weight, and distance between clips. Six radiation oncologists delineated the tumor bed volumes and rated the cavity visualization scores (CVS). The primary endpoint was the agreement between observers measured using a Conformity Index (Cx). RESULTS: Forty-two patients were included, 21 hydrogel procedures and 21 controls, resulting in 315 observer pairs. The feasibility of the intervention was 100%. The median Cx was higher in the intervention group (Cx = 0.70, IQR [0.54-0.79]) than in the control group (Cx = 0.54, IQR [0.42-0.66]), p < 0.00, as were the CVS (3.5 [2.5-4.5] versus 2.5 [2-3.5], p < 0.001). The rate of surgical site infections was similar to literature. CONCLUSIONS: The use of radiopaque PEG enables to identify the lumpectomy cavity, resulting in a high inter-observer agreement for radiotherapy target definition. This intervention is easy to perform and blend well into current practice.