151751-Najiba-Chargi

169 Surgery: skeletal muscle mass, arterial calcification and laryngectomy occurrence of adverse outcomes. Accurate identification of high-risk patients for PCF may provide an opportunity for preoperative interventions to decrease the risk. It seems unlikely to decrease the amount of arterial calcifications in the preoperative period, but preopera- tive optimization of general cardiovascular status or other risk factors associated with PCF which may co-exist might decrease the risk of a PCF. 27–29 Arterial calcifications as evidence for cardiovascular disease may warrant further examination and medical intervention prior to surgery. A surgical solution in high risk patients may be to use a pectoralis major overlay flap to reinforce the suture line of the neopharynx by covering it with healthy muscle and decrease the risk of PCF. 30 In reconstructive microsurgery, radiological evidence of atherosclerosis may also aid in choosing the optimal flap for recontruction. 31 There are several limitations that this study needs to address. It is apparent that relevant clinical data such as known cardiovascular disease and diabetes, was missing in our data- base due to missing information in particular in the earlier years of the study period. Also, some traditional cardiovascular risk factors such as serum cholesterol are missing, because these are not routinely measured at our clinic. Smoking and age was included in analysis, and the ASA classification was used as a surrogate for comorbidities, but we acknowledge that this provides limited information on specific comorbidities. 32 Recent studies do suggest that coronary arterial calcification scores or peripheral arterial calcification scores derived from CT imaging are reliable assessment methods for cardiovascular disease, and can identify patients at high risk that would not have been identified using traditional cardiovascular risk factors. 14,33,34 Second, a visual grading system for arterial calcification as opposed to calcium scores may lead to an observer bias and necessitate a learning curve. Automatic calcium scoring systems are not yet available using head and neck contrast en- hanced CT imaging, but research into automatic arterial calcification scoring on contrast-en- hanced CT imaging is ongoing, and this may in the future be available. 35,36 Moreover, machine learning and radiomics using CT features, e.g. skeletal muscle mass (sarcopenia), skeletal muscle quality and arterial calcification, from routinely performed CT imaging of the head and neck area may be helpful to identify patients at high risk for fistula formation after laryngec - tomy. In this study, all calcification scoring was performed by one observer: an experienced radiologist with a research interest and extensive experience with arterial calcification on CT imaging. The inter- and intraobserver variability was not researched in this study, but previ - ously found to be good in several studies also in less experienced observers. 18,37 Acknowledging these limitations, we do believe that this study provides a relevant novel ap- plication of routinely performed, readily available CT imaging of the head and neck area for optimization of the identification process of patients undergoing TL at high risk of developing a PCF. More research into the method of quantification of arterial calcification in head and neck cancer patients and its clinical application is warranted and clarification of its relevance for fistula prevention is needed. 9