Objectives. Prognosticating idiopathic sudden sensorineural hearing loss (ISSNHL) is an important challenge. In our study, a dataset was split into training and test sets and cross-validation was implemented on the training set, thereby determining the hyperparameters for machine learning models with high test accuracy and low bias. The effectiveness of the following five machine learning models for predicting the hearing prognosis in patients with ISSNHL after 1 month of treatment was assessed: adaptive boosting, K-nearest neighbor, multilayer perceptron, random forest (RF), and support vector machine (SVM).Methods. The medical records of 523 patients with ISSNHL admitted to Korea University Ansan Hospital between January 2010 and October 2017 were retrospectively reviewed. In this study, we analyzed data from 227 patients (recovery, 106; no recovery, 121) after excluding those with missing data. To determine risk factors, statistical hypothesis tests (e.g., the two-sample t-test for continuous variables and the chi-square test for categorical variables) were conducted to compare patients who did or did not recover. Variables were selected using an RF model depending on two criteria (mean decreases in the Gini index and accuracy).Results. The SVM model using selected predictors achieved both the highest accuracy (75.36%) and the highest F-score (0.74) on the test set. The RF model with selected variables demonstrated the second-highest accuracy (73.91%) and F-score (0.74). The RF model with the original variables showed the same accuracy (73.91%) as that of the RF model with selected variables, but a lower F-score (0.73). All the tested models, except RF, demonstrated better performance after variable selection based on RF.Conclusion. The SVM model with selected predictors was the best-performing of the tested prediction models. The RF model with selected predictors was the second-best model. Therefore, machine learning models can be used to predict hearing recovery in patients with ISSNHL.

The increasing use of personal listening devices (PLDs) has been accompanied by a rise in the prevalence of hearing loss (HL) in younger age groups. However, there have been few reports on the relationship between HL and leisure noise exposure (LNE) in adolescents. The purpose of our study was to investigate the prevalence of HL in students attending general middle and high schools and to identify factors affecting HL prevalence. A total of 2,879 nationally representative adolescents in the first years of middle and high school underwent audiometric testing and otological examinations, and completed questionnaires, from June to December 2016. A speech-frequency hearing loss (SFHL) was considered present when the pure tone averages (PTAs) at 0.5, 1, and 2 kHz were ≥ 15 dB and a high-frequency hearing loss (HFHL) was considered present when the PTAs at 3, 4, 6, and 8 kHz were ≥ 15 dB. About 17% of Korean adolescents exhibited at least slight HL. The prevalence rates of SFHL and HFHL in the poorer ear were 11.6% and 10.3%, respectively, among Korean adolescents. The use of local area network (LAN) gaming centers and an experience of being asked by others to lower earphone volume were associated with both SFHL and HFHL. It is important to avoid excessive LNE to prevent adolescent HL. Additionally, SFHL or HFHL in the poorer ear was associated with lower academic performance.

To evaluate long-term hearing changes and surgical outcomes for middle ear cholesteatoma with labyrinthine fistula following complete surgical removal of the cholesteatoma matrix. Consecutive samples of 43 patients who underwent 1-stage complete resection of the cholesteatoma matrix overlying a labyrinthine fistula were obtained. Immediate and long-term hearing changes were analyzed in association with fistula size. Accuracy of various diagnostic examinations was assessed. Immediately postsurgery, the average bone conduction threshold (43.34 dB) did not differ significantly from the preoperative value (36.66 dB, P = .083). There were also minimal changes thereafter (45.63 dB) without recurrent case over an average follow-up time of 38.3 months (range, 17-69 months). More than 10 dB hearing loss was found in 7 patients with a case of intraoperative perilymph leakage (2.3%), although 5 of them had had preoperative air conduction threshold above 90 dB. Their fistulas were significantly larger than those of patients without hearing loss (P = .027). Although caution is required for total removal of a large fistula, owing to increased risk of postoperative hearing deterioration, 1-stage complete resection of cholesteatoma matrix on labyrinthine fistula could be effective in disease control and long-term hearing preservation.

Objectives. The impacts of ventilation tube (VT) type and effusion composition on the VT extrusion rate and complications in children with otitis media remain unclear. This part II study evaluated the factors affecting the extrusion rate, recurrence rate, and complications of VT insertion.Methods. A prospective study was conducted between June 2014 and December 2016 (the EVENT study [analysis of the effectiveness of ventilation tube insertion in pediatric patients with chronic otitis media]), with follow-up data collected until the end of 2017. Patients aged <15 years diagnosed with otitis media with effusion who received VT insertion were recruited at 15 tertiary hospitals. The primary outcomes were time to extrusion of VT, time to effusion recurrence, and complications.Results. Data from 401 patients were analyzed. After excluding the results of long-lasting tubes (Paparella type II and T-tubes), silicone tubes (Paparella type I) exhibited a significantly longer extended time to extrusion (mean, 400 days) than titanium tubes (collar-button-type 1.0 mm: mean, 312 days; P<0.001). VT material (hazard ratio [HR], 2.117, 95% confidence interval [CI], 1.254–3.572; P=0.005), age (HR, 3.949; 95% CI, 1.239–12.590; P=0.02), and effusion composition (P=0.005) were significantly associated with the time to recurrence of middle ear effusion. Ears with purulent (mean, 567 days) and glue-like (mean, 588 days) effusions exhibited a shorter time to recurrence than ears with serous (mean, 846 days) or mucoid (mean, 925 days) effusions. The revision VT rates during follow-up were 3.5%, 15.5%, 10.4%, and 38.9% in ears with serous, mucoid, glue-like, and purulent effusions, respectively (P<0.001). The revision surgery rates were higher among patients aged <7 years than among those aged ≥7 years.Conclusion. Silicone tubes (Paparella type I) were less prone to early extrusion than titanium 1.0 mm tubes. VT type, patient age, and effusion composition affected the time to recurrence of effusion.

OBJECTIVES: The aim of this study was to evaluate the preoperative and postoperative facial nerve (FN) function in patients with FN schwannoma (FNS) and analyze the duration of preoperative facial palsy according to the preoperative and postoperative facial function. MATERIALS and METHODS: We retrospectively reviewed the medical records of 29 patients with FNS who underwent surgery. We evaluated the FN function according to the type of FN manipulation and location of the anastomoses in the cable nerve graft, and we also analyzed the duration of facial palsy according to the facial function before and after surgery. RESULTS: All 4 patients who underwent nerve-stripping surgery had the House–Brackmann (H-B) Grade III, 12 of 21 who underwent a cable nerve graft had the H-B Grade III or better postoperatively, and all 4 who underwent a hypoglossal facial crossover had the H-B Grade IV. Patients who underwent cable nerve grafting were more likely to have better FN function when the proximal anastomosis site was located in the internal auditory canal, geniculate ganglion, tympanic segment of FN, and distal end in the mastoid segment of FN. The duration of preoperative facial palsy was statistically shorter in patients with better postoperative facial function. CONCLUSION: Surgery can be considered in patients with FNS who have the H-B Grade III or worse. A shorter duration of facial palsy prior to surgery resulted in better postoperative facial function. Cite this article as: Kang WS, Han JJ, Rhee R, Lee JH, Koo JW, Chung JW. Surgical Outcomes of Intratemporal Facial Nerve Schwannomas According to Facial Nerve Manipulation. J Int Adv Otol 2019; 15(3): 415-9.