BACKGROUND: To develop a deep learning(DL) model utilizing ultrasound images, and evaluate its efficacy in distinguishing between benign and malignant parotid tumors (PTs), as well as its practicality in assisting clinicians with accurate diagnosis.
METHODS: A total of 2211 ultrasound images of 980 pathologically confirmed PTs (Training set: n = 721; Validation set: n = 82; Internal-test set: n = 89; External-test set: n = 88) from 907 patients were retrospectively included in this study. The optimal model was selected and the diagnostic performance evaluation is conducted by utilizing the area under curve (AUC) of the receiver-operating characteristic(ROC) based on five different DL networks constructed at varying depths. Furthermore, a comparison of different seniority radiologists was made in the presence of the optimal auxiliary diagnosis model. Additionally, the diagnostic confusion matrix of the optimal model was calculated, and an analysis and summary of misjudged cases\' characteristics were conducted.
RESULTS: The Resnet18 demonstrated superior diagnostic performance, with an AUC value of 0.947, accuracy of 88.5%, sensitivity of 78.2%, and specificity of 92.7% in internal-test set, and with an AUC value of 0.925, accuracy of 89.8%, sensitivity of 83.3%, and specificity of 90.6% in external-test set. The PTs were subjectively assessed twice by six radiologists, both with and without the assisted of the model. With the assisted of the model, both junior and senior radiologists demonstrated enhanced diagnostic performance. In the internal-test set, there was an increase in AUC values by 0.062 and 0.082 for junior radiologists respectively, while senior radiologists experienced an improvement of 0.066 and 0.106 in their respective AUC values.
CONCLUSIONS: The DL model based on ultrasound images demonstrates exceptional capability in distinguishing between benign and malignant PTs, thereby assisting radiologists of varying expertise levels to achieve heightened diagnostic performance, and serve as a noninvasive imaging adjunct diagnostic method for clinical purposes.

UNASSIGNED: Parotidectomy increases childhood challenges. This study aimed to determine the clinical profiles, investigations, and outcomes of pediatric patients who had undergone parotidectomy.
UNASSIGNED: A multicenter retrospective review of parotidectomy in pediatric patients between 2007 and 2020.
UNASSIGNED: In 108 parotidectomies, the final diagnoses were benign (47.22%), malignant (36.11%), and non-neoplastic (16.67%). The incidence of facial palsy was 37.03%, which was significantly lower in the superficial group than that in the total parotidectomy group (p = .021). The incidence of facial nerve palsy was significantly higher in the malignancy group than that in the benign group (p = .035). Magnetic resonance imaging (MRI) detected malignancy with 92.8% overall accuracy, 83.3% sensitivity, and 100% specificity. The sensitivity and specificity of fine-needle aspiration (FNA) were 54.2% and 92.7%, respectively.
UNASSIGNED: Parotidectomy is commonly performed for benign and non-neoplastic diseases in pediatric patients. Facial nerve palsy is significantly associated with malignant tumors and total parotidectomy. MRI is the most accurate imaging modality for diagnosing malignant lesions. FNA exhibits moderate agreement with the final pathology.
UNASSIGNED: Level IV.

Parotidectomy is the treatment of choice for benign tumors of the parotid gland, with the modified Blair incision most commonly utilized. This retrospective analysis aimed to determine the incidence of complications and assess the relationship between the mass and scar characteristics, in patients who had undergone parotidectomy.
The scar characteristics of patients who had undergone parotidectomy for benign neoplasms at our medical center between 2013 and 2019 were evaluated.
Overall, 49 patients met the inclusion criteria, of whom 33 agreed to participate in the study. The mean patient age was 52.8 years; 57.5% were males. No correlation was identified between the parotid mass size or location and scar type or length. The most common complications in this study were similar to those reported in the literature.
The results of this study suggest that the smaller, tailored modified Blair incision for parotidectomy does not increase complication rates. Future studies with larger cohorts should be conducted to further assess the potential benefits of tailored incision size during modified Blair parotidectomies.

OBJECTIVE: To evaluate the effectiveness of machine learning models based on morphological magnetic resonance imaging (MRI) radiomics in the classification of parotid tumors.
METHODS: In total, 298 patients with parotid tumors were randomly assigned to a training and test set at a ratio of 7:3. Radiomics features were extracted from the morphological MRI images and screened using the Select K Best and LASSO algorithm. Three-step machine learning models with XGBoost, SVM, and DT algorithms were developed to classify the parotid neoplasms into four subtypes. The ROC curve was used to measure the performance in each step. Diagnostic confusion matrices of these models were calculated for the test cohort and compared with those of the radiologists.
RESULTS: Six, twelve, and eight optimal features were selected in each step of the three-step process, respectively. XGBoost produced the highest area under the curve (AUC) for all three steps in the training cohort (0.857, 0.882, and 0.908, respectively), and for the first step in the test cohort (0.826), but produced slightly lower AUCs than SVM in the latter two steps in the test cohort (0.817 vs. 0.833, and 0.789 vs. 0.821, respectively). The total accuracies of XGBoost and SVM in the confusion matrices (70.8% and 59.6%) outperformed those of DT and the radiologist (46.1% and 49.2%).
CONCLUSIONS: This study demonstrated that machine learning models based on morphological MRI radiomics might be an assistive tool for parotid tumor classification, especially for preliminary screening in absence of more advanced scanning sequences, such as DWI.
CONCLUSIONS: • Machine learning algorithms combined with morphological MRI radiomics could be useful in the preliminary classification of parotid tumors. • XGBoost algorithm performed better than SVM and DT in subtype differentiation of parotid tumors, while DT seemed to have a poor validation performance. • Using morphological MRI only, the XGBoost and SVM algorithms outperformed radiologists in the four-type classification task for parotid tumors, thus making these models a useful assistant diagnostic tool in clinical practice.

OBJECTIVE: This study investigated the characteristics, diagnosis, and treatment of Warthin tumors (WTs) to explore the possibility of managing patients by observation.
METHODS: We reviewed the records of 1167 patients with benign parotid tumors who were seen in our department between September 1999 and April 2021. Among them, 387 cases were WT and 668 cases were pleomorphic adenoma. We evaluated preoperative diagnoses of WT by symptoms/signs, fine-needle aspiration cytology (FNAC), imaging, such as ultrasonography and magnetic resonance imaging, and technetium-99m pertechnetate (Tc-99m) scintigraphy. Fisher\'s exact test and the Mann-Whitney U test were used in statistical analyses.
RESULTS: Warthin tumors were treated by surgery in 238 cases and follow-up in 149 cases. The 238 patients were diagnosed as WT at the final pathology after surgery. Among them, 172 patients (72.3%) were determined as benign histological type by preoperative FNAC; in these 172 patients, 170 (71.4%) were correctly diagnosed as WT in the final pathology. Preoperative Tc-99m scintigraphy was performed in 69 patients diagnosed with WT by final pathology or FNAC, and the positive rate of Tc-99m scintigraphy in WT was 75.4%.
CONCLUSIONS: Combining FNAC and Tc-99m scintigraphy, as well as considering clinical findings, enables the diagnosis of WT in most cases. In particular, WT is more common in the elderly, grows more slowly, and is less likely to be malignant. Therefore, if WT can be diagnosed preoperatively with a high rate of correct diagnosis, it could be an accurate and effective means of managing patients through follow-up without surgery.

We investigated whether a fibrinogen-thrombin collagen sponge patch reduces postoperative complications of parotid gland surgery. This single-blinded, randomized controlled study included 165 patients who underwent parotid surgery for benign tumors (2018-2019) at a tertiary center. Primary outcomes were postoperative drain amount, days until drain removal, and discharge. Patients were scheduled for follow-up at 1 and 4 weeks, and 3 months after surgery. Complications including surgical site infection, pain, seroma, sialocele, salivary fistula, facial nerve palsy, Frey\'s syndrome with subjective symptoms, and facial asymmetry were analyzed. After identifying confounding variables, multivariate approaches were used. Histologic analysis was performed in a mouse model of salivary gland surgery. In total, 162 patients (77, fibrinogen-thrombin collagen patch group; 85, controls) were included, with no significant between-group differences other than resected tissue. Among postoperative total drain amount and days until drain removal and discharge, the only postoperative total drain was significantly lower in the patch group than in the control group in the adjusted model. Additionally, although validation through robust trials with longer follow-up is needed, we found the potential benefit of the fibrinogen patch on Frey\'s syndrome and facial asymmetry. In conclusion, fibrinogen-thrombin-impregnated collagen patches in parotidectomy can reduce postoperative drainage and improve outcomes.

UNASSIGNED: Close margin is a frequent situation after parotidectomy. The need for systematic prophylactic revision surgery is a question that arises regularly for malignant tumors, as it exposes to a high risk of facial palsy, while oncological benefits are unclear.
UNASSIGNED: retrospective study.
UNASSIGNED: Multicentric.
UNASSIGNED: We included all patients operated for systematic revision surgery in case of close margins after parotidectomy for a malignant tumor and analyzed the rate of tumor residue and its risk factors.
UNASSIGNED: A tumor residue was identified in 43.5% of 23 cases, but none in case of initial complete excision with supra-millimetric margins. Invaded lymph nodes were identified in 6 cases, but none in case of low-grade tumors.
UNASSIGNED: Systematic revision seems mandatory in case of infra-millimetric margins and high-grade tumors or positive lymph node; further studies are needed to confirm whether it can be spared for T1-T2/N0 low-grade tumors, with close margins but complete initial excision.

Salivary fistula is a relatively common complication in patients who have undergone a parotidectomy. The purpose of this study was to investigate the effects of bipolar coagulation forceps use on salivary fistulas.
From March 2015 to June 2020, 177 patients who underwent a parotidectomy in the Department of Oral and Maxillofacial Surgery at the Second Xiangya Hospital of Central South University were recruited. The patients were divided into an experimental group and a control group based on whether bipolar coagulation forceps or sutures were used, respectively.
The drainage output of the experimental group was significantly lower than that of the control group (p = 0.04). The duration of dressing pressure applied in the experimental group was significantly shorter than that in the control group (p = 0.0003). Moreover, the incidence of salivary fistula in the experimental group (9.8%, 8/82) was notably lower than that in the control group (34.7%, 33/95) (p < 0.0001). In the logistic regression model for salivary fistula development, both the use of bipolar coagulation forceps (p = 0.0021) and drainage output (p = 0.0237) were associated with the presence of salivary fistulas.
Our findings indicate that the use of bipolar coagulation forceps decreases the incidence of salivary fistula in patients who have undergone a parotidectomy. The use of bipolar coagulation forceps is a safe, effective, and convenient method to prevent salivary fistulas in patients who undergo a parotidectomy.
Current Controlled Trials ChiCTR2100044722, Date: 26/03/2021, Retrospectively registered.

Malignant parotid tumors account for approximately 20% of all parotid lesions. In addition to the various primary parotid lesions there are secondary parotid malignancies, such as metastases or lymphomas. Data on histopathological distribution of all malignancies-including secondary parotid lesions-is limited. Recent evidence indicated a rising surgical incidence of secondary parotid malignancies. This study aims to review the distribution of malignancies in parotid resections from a salivary gland center. A retrospective review of prospectively collected data for all patients who had received parotidectomy between 2014 and 2019 was performed. Histopathological distribution was displayed separately for all parotid malignancies and for primary parotid malignancies. Further, patients` characteristics were compared between benign and malignant parotid lesions and between the two most common malignant parotid lesions. Out of 777 patients, 614 (78.9%) patients had a benign and 164 (21.1%) patients had a malignant parotid lesion. The most common parotid malignancy was metastatic cutaneous squamous cell carcinoma (cSCC) accounting for 35.4% of all parotid malignancies. 71.5% of all malignant lesions were secondary malignancies. Patients with metastatic cSCC were significantly older (p < 0.001) and significantly more likely to be male (p < 0.001) than patients with primary parotid malignancies. No significant difference was found when the lesion size of metastatic cSCC was compared to primary parotid malignancies (p = 0.216). The present study shows the high prevalence of secondary parotid malignancies in patients who had received parotidectomy. Furthermore, it confirms a rising surgical incidence of metastatic cSCC to the parotid gland in a series from a salivary gland center. At this time, parotid surgery for malignant lesions is more likely to be performed for metastases than for primary parotid malignancies.

OBJECTIVE: We examined the incidence, risk factors, and recovery time for postoperative facial nerve paralysis in patients with benign parotid tumors and also investigated the usefulness of intraoperative facial nerve monitoring (FNM).
METHODS: Subjects were 902 patients with benign parotid tumors. Univariate and multivariate analyses were conducted for risk factors of postoperative facial nerve paralysis. We investigated the relationship between intraoperative FNM and postoperative facial nerve paralysis, and the recovery time for postoperative facial nerve paralysis according to tumor site.
RESULTS: Postoperative facial nerve paralysis occurred in 179 (19.8%) of 902 patients. Among them, 15.1%, 15.8%, and 37.5% had tumors in the superficial lobe, lower pole, and deep lobe, respectively. Paralysis risk factors were deep tumors, large tumors, long operation times, extensive bleeding, and non-use of FNM. Multivariate analysis determined female patients, deep tumors, and long operation times as significant risk factors, and female patients and deep tumors had an odds ratio of nearly 2. Use of intraoperative FNM resulted in a significantly lower incidence of facial nerve paralysis and was particularly useful in patients with superficial lobe tumors. Time to recovery from facial nerve paralysis was 6 months in 88% of the patients.
CONCLUSIONS: The incidence of paralysis should be kept as low as possible to enhance the postoperative quality of life of patients. The use of intraoperative FNM significantly lowered the incidence of paralysis in female patients with superficial tumors.