Grading criteria on radionecrosis of skin (GCRNS) in adult patients: Insights from Delphi study and validation process.

Zhu, Meng-Yue; Yang, Fei; Xie, Wei-Guo; Xi, Mao-Mao; Liu, Shu-Hua; Zhang, Wei; Wang, De-Yun; Zhang, Wei-Dong et al. · Burns · 2025

other · Level V

Where this comes from

Abstract

Radionecrosis of skin (RNS) is the severe consequence of radiodermatitis (RD), which is a common complication experienced in up to 95 % of cancer patients after radiation therapy (RT). Recent publications revealed an absence of grading criteria for RNS and only crude forms for RD. However, treatments for RNS of different severity vary widely in terms of treatment difficulty and prognosis. This Delphi study aims to provide wound repair surgeons and oncologists a comprehensive guide for accurate RNS diagnosis, facilitating more effective surgical or operative interventions. A three-round Delphi method was conducted between May and September 2022, involving 24 experts who specialize in the care or research of RNS. The process identified and weighted primary and secondary items to create a scale that reflects the severity of RNS. A receiver operating characteristic (ROC) curve was generated using the expert's ratings (n = 15) as the benchmark to establish the grading criteria on RNS (GCRNS) with patient data (n = 64). To assess test-retest reliability, 32 physicians re-evaluate 64 patients' files 2 weeks after their initial evaluation. The correlation between expert ratings and physician's re-evaluated grading was calculated to confirm concurrent validity. The final GCRNS consisted of 18 secondary items categorized under 6 primary items, including ulcer duration, peripheral skin, ulcer area, underlying disease, ulcer depth, and severe complications. Severity classification thresholds were proposed by ROC curve, defining severity levels as mild, moderate, and severe. Six primary items were positively correlated the severity of RNS with affirming its concurrent validity (r = 0.751, p < 0.001). Good internal consistency (α=0.831) and test-retest reliability (r = 0.969, p < 0.001) were demonstrated. GCRNS is an expert-driven and comprehensive approach to RNS assessment for further wound repair. By improving diagnostic accuracy, particularly among junior surgeons, GCRNS with strong concurrent validity and reliability, enhances clinical decision-making and supports more effective treatment planning.

Medical subject headings