Parotid Gland Tumors: 2-Year Prospective Clinicopathological Study

Development of a Machine Learning Algorithm to Forecast the Likelihood of Postoperative Neurological Complications in Patients With Parotid Tumors

Objective: The objective of this study was to create and verify a machine learning-driven predictive model to forecast the likelihood of facial nerve impairment in patients with parotid tumors following surgery. Methods: We retrospectively collected data from patients with parotid tumors between 2013 and 2023 to develop a prediction model for postoperative facial nerve dysfunction using 5 ML techniques: Logistic Regression (Logit), Random Forest (RF), XGBoost (XGB), Artificial Neural Network (ANN), and Support Vector Machine (SVM). Predictor variables were screened using binomial-LASSO regression. Results: The study had a total of 403 participants, out of which 56 individuals encountered facial nerve damage after the surgery. By employing binomial-LASSO regression, we have successfully identified 8 crucial predictive variables: tumor kind, tumor pain, surgeon's experience, tumor volume, basophil percentage, red blood cell count, partial thromboplastin time, and prothrombin time. The models utilizing ANN and Logit achieved higher area under the curve (AUC) values, namely 0.829, which was significantly better than the SVM model that had an AUC of 0.724. There were no noticeable disparities in the AUC values between the ANN and Logit models, as well as between these models and other techniques like RF and XGB. Conclusion: Using machine learning, our prediction model accurately predicts the likelihood that patients with parotid tumors may experience facial nerve damage following surgery. By using this model, doctors can assess patients' risks more accurately before to surgery, and it may also help optimize postoperative treatment techniques. It is anticipated that this tool would enhance patients' quality of life and therapeutic outcomes.

MR-Neurography of the facial nerve in parotid tumors: intra-parotid nerve visualization and surgical correlation

PURPOSE

One of the most severe complications in surgery of parotid tumors is facial palsy. Imaging of the intra-parotid facial nerve is challenging due to small dimensions. Our aim was to assess, in patients with parotid tumors, the ability of high-resolution 3D double-echo steady-state sequence with water excitation (DE3D-WE) (1) to visualize the extracranial facial nerve and its tracts, (2) to evaluate their relationship to the parotid lesion and (3) to compare MRI and surgical findings.

METHODS

A retrospective study was conducted including all patients with parotid tumors, who underwent MRI from April 2022 to December 2023. Two radiologists independently reviewed DE3D-WE images, assessing quality of visualization of the facial nerve bilaterally and localizing the nerve's divisions in relation to the tumor. MRI data were compared with surgical findings.

RESULTS

Forty consecutive patients were included (M:F = 22:18; mean age 56.3 ± 17.4 years). DE3D-WE could excellently visualize the nerve main trunk and the temporofacial division in all cases. The cervicofacial branch was visible in 99% of cases and visibility was good. Distal divisions were displayed in 34% of cases with a higher visibility on the tumor side (p < 0.05). Interrater agreement was high (weighted kappa 0.94 ± 0.01 [95% CI 0.92-0.97]). Compared to surgery accuracy of MRI in localizing the nerve was 100% for the main trunk, 96% for the temporofacial and 89% for the cervicofacial branches.

CONCLUSIONS

Facial nerve MR-neurography represents a reliable tool. DE3D-WE can play an important role in surgical planning of patients with parotid tumors, reducing the risk of nerve injury.

Deep learning for the automatic detection and segmentation of parotid gland tumors on MRI

OBJECTIVES

Parotid gland tumors (PGTs) often occur as incidental findings on magnetic resonance images (MRI) that may be overlooked. This study aimed to construct and validate a deep learning model to automatically identify parotid glands (PGs) with a PGT from normal PGs, and in those with a PGT to segment the tumor.

MATERIALS AND METHODS

The nnUNet combined with a PG-specific post-processing procedure was used to develop the deep learning model trained on T1-weighed images (T1WI) in 311 patients (180 PGs with tumors and 442 normal PGs) and fat-suppressed (FS)-T2WI in 257 patients (125 PGs with tumors and 389 normal PGs), for detecting and segmenting PGTs with five-fold cross-validation. Additional validation set separated by time, comprising T1WI in 34 and FS-T2WI in 41 patients, was used to validate the model performance.

RESULTS AND CONCLUSION

To identify PGs with tumors from normal PGs, using combined T1WI and FS-T2WI, the deep learning model achieved an accuracy, sensitivity and specificity of 98.2% (497/506), 100% (119/119) and 97.7% (378/387), respectively, in the cross-validation set and 98.5% (67/68), 100% (20/20) and 97.9% (47/48), respectively, in the validation set. For patients with PGTs, automatic segmentation of PGTs on T1WI and FS-T2WI achieved mean dice coefficients of 86.1% and 84.2%, respectively, in the cross-validation set, and of 85.9% and 81.0%, respectively, in the validation set. The proposed deep learning model may assist the detection and segmentation of PGTs and, by acting as a second pair of eyes, ensure that incidentally detected PGTs on MRI are not missed.

Clinico-histopathological review of 255 patients who underwent parotidectomy for pleomorphic adenoma: a 10-year retrospective study-a proposal for an optimal diagnostic and therapeutic algorithm for patients with recurrent pleomorphic adenoma

PURPOSE

Pleomorphic adenoma (mixed tumor) is the most common neoplasm of the parotid gland and one of the most frequent types of salivary gland tumor, generally with benign behavior and relatively slow growing. The adenomas could arise from the superficial, deep or from both superficial and deep parotid's lobes.

METHODS

The aim of this review is to retrospectively analyze the surgical management of patients with pleomorphic adenoma of the parotid gland performed at the Department of Otorhinolaryngology (Department of Sense Organs of "Azienda Policlinico Umberto I" in Rome), from 2010 to 2020, with a focus on the percentage of recurrence and on the complication related to surgery to suggest an optimal diagnostic and therapeutic algorithm for patients with recurrent pleomorphic adenoma. The analysis of the complications observed in case of different surgical approaches was performed using the X² test.

RESULTS

The choice of a surgical approach (superficial parotidectomy-SP, total parotidectomy-TP, extracapsular dissection-ECD) depends on several elements, such as the location and the size of the adenoma, the availability of existing technical facilities and the professional experience of the surgeon. A transient facial palsy was present in 37.6%, 2.7% reported a permanent facial nerve palsy, 1.6% developed a salivary fistula, 1.6% a post-operative bleeding and 2.3% showed Frey Syndrome.

CONCLUSION

The surgical management of this benign lesion is required, even in asymptomatic cases, to prevent the progressive growing and to reduce the risk of malignant transformation. The goal of surgical excision is to obtain the complete resection to minimize the risk of tumor recurrence and avoiding facial nerve disability. Therefore, an accurate preoperative study of the lesion and the choice of the most appropriate surgical treatment are essential to minimize the rate of recurrence.

Comparison of various surgical incisions in parotidectomy: A systematic review and network meta-analysis

Background

This network meta-analysis aimed to comprehensively compare the operative and postoperative outcomes of different parotidectomy incisions.

Methods

Embase, PubMed, Web of Science, and Cochrane Central Register of Controlled Trials were searched up to April 2022. A complete Bayesian network meta-analysis was performed using the Markov Monte Carlo method in OpenBUGS.

Results

Seventeen studies with 1609 patients were included. Thirteen were retrospective cohort studies, three were prospective cohort studies, and one was a randomized controlled study. The quality of evidence was rated as very low in most comparisons. The incision satisfaction score of the modified facelift incision (MFI), retroauricular hairline incision (RAHI), V-shaped incision (VI) were higher than that of the modified Blair incision (MBI) (MBI vs. MFI: mean difference [MD] -1.39; 95% credible interval [CrI] -2.23, -0.57) (MBI vs. RAHI: MD -2.25; 95% CrI -3.40, -1.12) (MBI vs. VI: MD -2.58; 95% CrI -3.71, -1.46); the tumor size treated by VI was smaller than that by MBI (MD 5.15; 95% CrI 0.76, 9.38) and MFI (MD 5.16; 95% CrI 0.34, 9.86); and the risk of transient facial palsy in the MFI was lower than that in the MBI (OR 2.13; 95% CrI 1.28, 3.64). There were no differences in operation time, drainage volume, wound infection, hematoma, salivary complications, Frey syndrome, or permanent facial palsy between incision types.

Conclusion

The traditional MBI is frequently used for large tumor volumes, but the incision satisfaction score is low and postoperative complication control is poor. However, emerging incisions performed well in terms of incision satisfaction scores and control of complications. More randomized controlled trials are needed to compare the different parotidectomy incisions. Patients should be fully informed about the characteristics of each incision to make the most informed decision, along with the physician’s advice.

Systematic Review Registration

PROSPERO, identifier CRD42022331756

Facial Nerve Repair following Acute Nerve Injury

Background  Acute facial nerve iatrogenic or traumatic injury warrants rapid management with the goal of reestablishing nerve continuity within 72 hours. However, reconstructive efforts should be performed up to 12 months from the time of injury since facial musculature may still be viable and thus facial tone and function may be salvaged.

Methods  Data of all patients who underwent facial nerve repair following iatrogenic or traumatic injury were retrospectively collected and assessed. Paralysis etiology, demographics, operative data, postoperative course, and outcome were examined.

Results  Twenty patients underwent facial nerve repair during the years 2004 to 2019. Data were available for 16 of them. Iatrogenic injury was the common category ( n  = 13, 81%) with parotidectomy due to primary parotid gland malignancy being the common surgery ( n  = 7, 44%). Nerve repair was most commonly performed during the first 72 hours of injury ( n  = 12, 75%) and most of the patients underwent nerve graft repair ( n  = 15, 94%). Outcome was available for 12 patients, all of which remained with some degree of facial paresis. Six patients suffered from complete facial paralysis (50%) and three underwent secondary facial reanimation (25%). There were no major operative or postoperative complications.

Conclusion  Iatrogenic and traumatic facial nerve injuries are common etiologies of acquired facial paralysis. In such cases, immediate repair should be performed. For patients presenting with facial paralysis following previous surgery or trauma, nerve repair should be considered up to at least 6 months of injury. Longstanding paralysis is best treated with standard facial reanimation procedures.

Clinicopathological features of incidental parotid lesions

Background

The purpose of this study is to determine the histopathological spectrum and risk of primary malignancy of asymptomatic parotid lesions incidentally discovered on cross-sectional imaging.

Methods

Over a 10-year period, 154 patients underwent 163 ultrasound-guided parotid lesion biopsies at our institution. This retrospective chart review included 89 lesions in 87 patients with asymptomatic parotid lesions discovered on cross-sectional imaging studies performed for unrelated clinical indications. The histopathologic findings of all sampled lesions were reviewed. We evaluated the patient demographics and pathological diagnoses of sampled parotid lesions to determine the histopathological spectrum and risk of malignancy.

Results

The average age was 67.5 years and 92 % were males. 25 % of patients had bilateral lesions. The average size of the parotid lesions was 1.5 cm and 91 % were located in the superficial lobe. 92.1 % of lesions were benign with Warthin tumor being the most common diagnosis followed by pleomorphic adenoma. 2.3 % of lesions were primary parotid malignant neoplasms, while 5.6 % were metastatic lesions in patients with known malignancy.

Conclusions

The incidence of primary parotid malignant neoplasm in asymptomatic incidentally discovered parotid lesions is low. Imaging or clinical follow-up may be considered in patients with incidental parotid lesions who prefer to avoid biopsy.

Mucoepidermoid Carcinoma of the Parotid Gland Presenting as a Fungating Exophytic Mass: A Surgical Challenge Rooted in Extensive Nerve Encasement

Parotid glands, which are predominantly composed of serous acini, are the largest salivary glands in the human body. Mucoepidermoid carcinoma (MEC) of the parotid gland is the most common parotid tumour that routinely presents as a painless, fixed mass below the ears. However, its presentation as a fungating mass remains exceedingly rare. Due to the intimate anatomical relationship between the parotid gland and the facial nerve, parotid malignancies can culminate in facial nerve palsy, posing an onerous surgical challenge. In this paper, we chronicle the case of a male who presented with a fungating/exophytic mass and facial nerve weakness. A superficial parotidectomy was performed, and the eventual histopathological workup divulged an advanced mucoepidermoid parotid carcinoma entirely encasing the marginal mandibular nerve. Consequently, the marginal mandibular nerve was sacrificed, alluding to the remarkable surgical challenge encountered. Clinically, this manifested as an impairment of the motor function in the patient's left lower lip. 

Fine-Needle Aspiration Cytology Is an Effective Diagnostic Tool in Paediatric Patients with Mucoepidermoid Carcinoma as Secondary Neoplasm

BACKGROUND

Mucoepidermoid carcinoma (MEC) is the most common malignant salivary gland tumour in paediatric population, accounting for 16% of all cases. Patients affected by a previous solid or leukaemic neoplasm during their childhood may develop a second different tumour during the follow-up. In this setting, salivary gland MEC is relatively frequent, accounting for 6% of the second neoplasms in paediatric patients. Consequently, the occurrence of salivary gland nodules in paediatric patients with a previous neoplasm should be considered an event with a high risk of malignancy that poses peculiar diagnostic challenges.

SUMMARY

This study was designed to define clinical and instrumental findings and morphological features of MEC on fine-needle aspiration cytology (FNAC) samples in paediatric patients with and without a previous neoplasm. Five patients under 19 years are included in this series. FNAC was performed in all patients on a parotid nodule. We have identified 2 groups of patients: (a) 2 cases with previous history of malignancy (acute lymphoblastic leukaemia and Hodgkin lymphoma) and (b) 3 cases without previous malignant neoplasms. In all cases, a final diagnosis of MEC was rendered. Key Messages: MEC may occur as a second malignancy in paediatric patients. FNAC is certainly a valid and accurate diagnostic tool for this type of neoplasm, even in the paediatric age, allowing the correct management of the patients.

Review and characteristics of 585 salivary gland neoplasms from a tertiary hospital registered in the Polish National Major Salivary Gland Benign Tumors Registry over a period of 5 years: a prospective study

&lt;b&gt;Introduction:&lt;/b&gt; A Polish National Major Salivary Gland Benign Tumors Registry (SGR) is a report of benign salivary gland neoplasms (SGNs) from 26 different centres in Poland, introduced in 2014. The aim of this study is to analyze demographic characteristics and clinico-pathological factors of benign SGNs treated in large tertiary institutions and to determine possible correlations between selected variables. &lt;br&gt;&lt;b&gt;Material and method:&lt;/b&gt; Analysis of 585 patients recorded in SGR and operated on for SGNs in the Department of Otolaryngology and Laryngological Surgery, University of Medical Sciences, Poznań, Poland, over a 5-year period. Patient age, sex, occupation, place of residence, tumor location, size, histology, recurrence, facial nerve function after surgery, wound healing, surgery procedure, availability of pre-operative imaging examinations, fine-needle aspiration cytology (FNAC) results were analyzed. &lt;br&gt;&lt;b&gt;Results:&lt;/b&gt; 338 females and 247 males with a mean age of 53 years were operated on. In total, 96.2% of tumors originated from the parotid and 3.8% from the submandibular gland. The most frequent primary tumor diameter was 2-4 cm (59.5%) followed by &lt;2 cm (29.2%) and &gt;4 cm (8.4%). Tumors of over 4 cm were frequently removed by partial superficial parotidectomy, while those under 2 cm - by extracapsullar dissection (ECD). Pleomorphic adenomas (PA) were predominant (58.8%), followed by Warthin's tumor (WT) - 37.1%. Patients with WT were on average 13.3 years older than patients with PA. &lt;br&gt;&lt;b&gt;Discussion:&lt;/b&gt; This research can be helpful to better understand the clinico-pathological features of SGNs. Long-termin hospital-based analysis is important for subsequent metaanalyses and comparisons with other centers. The reasons why not all patients' data are reported to the national SGR should be further precisely analyzed.