Premium
In Response to A Nomogram to Predict Osteoradionecrosis in Oral Cancer After Marginal Mandibulectomy and Radiotherapy
Author(s) -
Tsai TsungYou,
Tay Ze Yun,
Chang KaiPing,
Huang YuTung
Publication year - 2019
Publication title -
the laryngoscope
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.181
H-Index - 148
eISSN - 1531-4995
pISSN - 0023-852X
DOI - 10.1002/lary.28092
Subject(s) - medicine , otorhinolaryngology , osteoradionecrosis , head and neck cancer , oral and maxillofacial surgery , head and neck surgery , general surgery , head and neck , radiation therapy , surgery
We would like to thank Dr. Collins for his comments regarding the methodological issues mentioned in his letter and the recommendation to consult the TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) statement based on their recent publication. At the same time, there are some misunderstandings that we believe need to be clarified. In our article, we only used five variables in our final nomogram, rather than 17 variables as indicated in the letter. Hence, the events per variable (EPV) number was three (15 patients with occurrence of osteoradionecrosis [ORN] while examining five variables) instead of 0.9. Although the criteria of EPV ≥10 is thought to be associated with better predictive performance, there is still some controversy regarding determining a good sample size based on this rule. Additionally, the wide confidence interval in the calibration plot was also noted; nonetheless, the lowest limit of the fraction ORN status (either 1 or 3 years) is higher than 0.8, which indicates that the prediction could still be valuable for the clinician as a reference. In our study, the 167 patients in the study cohort who received marginal mandibulectomy and postoperative radiation therapy were enrolled from 3,087 patients with oral squamous cell carcinoma (OSCC) undergoing surgical resection and free-tissue transfer during a 7-year period (2006–2012). It has to be emphasized that even in a highvolume academic referral center, the proportion of OSCC patients who underwent both marginal mandibulectomy and postoperative radiation therapy was still limited. The occurrence of osteoradionecrosis after the marginal mandibulectomy and radiotherapy was consequently even smaller (only 15 cases). Therefore, the reality of the small number of incident cases, despite a large study population, limit the chance of any internal or external validation for the current study. Moreover, the radiologic data of all of the OSCC patients who underwent marginal mandibulectomy have been thoroughly reviewed, and the current study is the first to propose the use of resected bone height to original bone height ratio ≥50% as the cutoff value and the associated nomogram to predict the occurrence of mandibular osteoradionecrosis. In our opinion, mandibular osteoradionecrosis is not a common complication after contemporary OSCC treatment, and our study results and recommendation should still be considered valuable based on this relatively large-scale retrospective study.