Neff, Brian A.

Objectives/HypothesisInverting papillomas (IPs) are benign locally invasive tumors that most commonly present within the sinonasal cavity. Temporal bone involvement is exceedingly rare, with fewer than 30 cases reported within the English literature to date.Study DesignCase series and systematic review of the literature.MethodsFour consecutive subjects with temporal bone inverting papilloma (TBIP) were treated, and an additional 28 previously published cases were identified in the literature. Main outcome measures were disease presentation, diagnostic evaluation, management strategy, and outcome.ResultsA total of 32 cases were analyzed. The median age at diagnosis was 54 years (mean 54.1; range 19–81 years). Nineteen (59%) patients had synchronous or metachronous sinonasal IP, whereas 13 (41%) had isolated temporal bone disease without sinus involvement. Over half of the patients undergoing microsurgical resection experienced at least one recurrence. Compared to patients with a history of sinus IP, subjects with primary TBIP were younger at time of presentation (44 vs. 58 years; P = 0.012); were more commonly female (62% vs. 32%; P = 0.15); and were less likely to have intracranial spread (8% vs. 26%; P = 0.36), cranial neuropathy (8% vs. 26%; P = 0.36), human papillomavirus positivity (11% vs. 57%; P = 0.11), or associated carcinoma (0% vs. 47%; P = 0.004).ConclusionsInverting papilloma of the lateral skull base is rare and can pose a significant therapeutic challenge. Primary lesions of the temporal bone appear to follow a less aggressive clinical course when compared to those arising in association with sinonasal disease. Gross total resection is the preferred method of treatment, when feasible, given the high rate of recurrence with subtotal resection and risk of associated malignancy.Level of Evidence4. Laryngoscope, 125:2576–2583, 2015

OBJECT The optimal treatment for sporadic vestibular schwannoma (VS) is highly controversial. To date, the majority of studies comparing treatment modalities have focused on a narrow scope of technical outcomes including facial function, hearing status, and tumor control. Very few publications have investigated health-related quality of life (HRQOL) differences between individual treatment groups, and none have used a disease-specific HRQOL instrument. METHODS All patients with sporadic small- to medium-sized VSs who underwent primary microsurgery, stereotactic radiosurgery (SRS), or observation between 1998 and 2008 were identified. Subjects were surveyed via postal questionnaire using the 36-Item Short Form Health Survey (SF-36), the 10-item Patient-Reported Outcomes Measurement Information System short form (PROMIS-10), the Glasgow Benefit Inventory (GBI), and the Penn Acoustic Neuroma Quality-of-Life (PANQOL) scale. Additionally, a pool of general population adults was surveyed, providing a nontumor control group for comparison. RESULTS A total of 642 respondents were analyzed. The overall response rate for patients with VS was 79%, and the mean time interval between treatment and survey was 7.7 years. Using multivariate regression, there were no statistically significant differences between management groups with respect to the PROMIS-10 physical or mental health dimensions, the SF-36 Physical or Mental Component Summary scores, or the PANQOL general, anxiety, hearing, or energy subdomains. Patients who underwent SRS or observation reported a better total PANQOL score and higher PANQOL facial, balance, and pain subdomain scores than the microsurgical cohort (p < 0.02). The differences in scores between the nontumor control group and patients with VS were greater than differences observed between individual treatment groups for the majority of measures. CONCLUSIONS The differences in HRQOL outcomes following SRS, observation, and microsurgery for VS are small. Notably, the diagnosis of VS rather than treatment strategy most significantly impacts quality of life. Understanding that a large number of VSs do not grow following discovery, and that intervention does not confer a long-term HRQOL advantage, small- and medium-sized VS should be initially observed, while intervention should be reserved for patients with unequivocal tumor growth or intractable symptoms that are amenable to treatment. Future studies assessing HRQOL in VS patients should prioritize use of validated disease-specific measures, such as the PANQOL, given the significant limitations of generic instruments in distinguishing between treatment groups and tumor versus nontumor subjects.

Objectives/HypothesisTo investigate the influence of posttreatment audiovestibular symptoms, facial neuropathy, and headache on long‐term quality‐of‐life outcomes in patients with sporadic vestibular schwannoma (VS) utilizing the Short Form 36 (SF‐36) Health Survey and the Penn Acoustic Neuroma Quality of Life (PANQOL) scale.Study DesignCross‐sectional observation study.MethodsPatients with sporadic small‐ or medium‐sized VS (< 3.0 cm) who were evaluated between 1998 and 2008 at two independent tertiary academic referral centers were surveyed. Multivariable associations with the PANQOL total score and the SF‐36 physical and mental component scores evaluated using regression analysis.ResultsA total of 538 surveyed patients returned a completed questionnaire, providing a response rate of 79%. Two hundred forty‐seven (46%) patients underwent stereotactic radiosurgery, 143 (27%) microsurgery, and 148 (28%) observation. Multivariable regression analysis revealed that ongoing dizziness was associated with the greatest reduction in PANQOL total score, followed by headache. After adjusting for all examined features, ongoing dizziness and ongoing headache were the only two variables that were associated with both the SF‐36 physical and mental component scores. Patient sex and treatment modality did not significantly influence PANQOL or SF‐36 scores.ConclusionsOngoing dizziness and headache are the strongest predictors of long‐term quality‐of‐life reduction in patients with sporadic VS, while the impact of hearing loss, facial nerve function, and tinnitus are less by comparison. This information may be valuable for patient counseling, refinement of VS quality‐of‐life assessment instruments, and determining high‐yield targets for therapy in efforts to further improve patient outcomes.Level of Evidence4. Laryngoscope, 125:1697–1702, 2015

Objectives/HypothesisFibrous dysplasia is a condition of nonmalignant osseous change and may occur in a monostotic or polyostotic pattern, the latter potentially being associated with McCune‐Albright syndrome. Symptoms are highly variable and dependent upon lesion location and size.Study DesignRetrospective review.MethodsConsecutive subjects with fibrous dysplasia of the temporal bone were evaluated between 2000 and 2013 at two tertiary academic referral centers. Main outcome measures included disease presentation, diagnostic evaluation, management strategy, and outcome.ResultsSixty‐six patients with fibrous dysplasia of the skull were found to have involvement of the temporal bone. The mean age at diagnosis was 25 years, 39 (59%) were female, and the mean duration of follow‐up was 48 months. Six (11%) patients had monostotic disease, with the remaining 60 (89%) patients having the polyostotic form; 16 (24%) patients had McCune‐Albright syndrome. The most common presenting complaint was headache (59%), followed by hearing loss (29%). The most common exam finding was cosmetic deformity (50%). Cholesteatoma (3%) and spontaneous cerebrospinal fluid fistula (1.5%) were found in a small percentage. No patients had evidence of motor cranial neuropathy by history or physical examination.ConclusionsThe clinical presentation of fibrous dysplasia involving the temporal bone is variable. A growing number of patients are diagnosed incidentally through imaging, and since most patients experience a benign course, the majority can be followed clinically without need for intervention.Level of Evidence4. Laryngoscope, 125:1438–1443, 2015

OBJECT The following study was conducted to identify risk factors for a postoperative CSF leak after vestibular schwannoma (VS) surgery. METHODS The authors reviewed a prospectively maintained database of all patients who had undergone resection of a VS at the Mayo Clinic between September 1999 and May 2013. Patients who developed a postoperative CSF leak within 30 days of surgery were compared with those who did not. Data collected included patient age, sex, body mass index (BMI), tumor size, tumor side, history of prior tumor treatment, operative time, surgical approach, and extent of resection. Both univariate and multivariate regression analyses were performed to evaluate all variables as risk factors of a postoperative CSF leak. RESULTS A total of 457 patients were included in the study, with 45 patients (9.8%) developing a postoperative CSF leak. A significant association existed between increasing BMI and a CSF leak, with those classified as overweight (BMI 25–29.9), obese (BMI 30–39.9), or morbidly obese (BMI ≥ 40) having a 2.5-, 3-, and 6-fold increased risk, respectively. Patients undergoing a translabyrinthine (TL) approach experienced a higher rate of CSF leaks (OR 2.5, 95% CI 1.3–4.6; p = 0.005), as did those who had longer operative times (OR 1.04, 95% CI 1.02–1.07; p = 0.0006). The BMI, a TL approach, and operative time remained independent risk factors on multivariate modeling. CONCLUSIONS Elevated BMI is a risk factor for the development of a postoperative CSF leak following VS surgery. Recognizing this preoperatively can allow surgeons to better counsel patients regarding the risks of surgery as well as perhaps to alter perioperative management in an attempt to decrease the likelihood of a leak. Patients undergoing a TL approach or having longer operative times are also at increased risk of developing a postoperative CSF leak.

Objective (1) To characterize long‐term dizziness following observation, microsurgery, and stereotactic radiosurgery (SRS) for small to medium‐sized vestibular schwannoma (VS) using a validated self‐assessment inventory; and (2) to identify clinical variables associated with long‐term dizziness handicap. Study Design Cross‐sectional observational study. Setting Two independent tertiary academic referral centers: one located in the United States and one in Norway. Subjects and Methods All patients with sporadic VS of less than 3 cm who underwent primary microsurgery, SRS, or observation between 1998 and 2008 were identified. Subjects were surveyed via a postal questionnaire using the Dizziness Handicap Inventory (DHI) and a VS symptom questionnaire. Results The overall survey response rate was 79%. A total of 538 respondents (mean age, 64 years; 56% female) were analyzed, and the mean time interval between treatment and survey was 7.7 years. Pretreatment variables associated with greater dizziness handicap included female sex, older age, larger tumor size, preexisting diagnosis of headache or migraine, and symptoms of dizziness predating treatment. Significant posttreatment features strongly associated with poor long‐term DHI scores included frequency and severity of ongoing headache. On multivariable analysis, treatment modality did not influence long‐term dizziness handicap. Conclusion At a mean of approximately 8 years following treatment, over half of patients with VS reported ongoing dizziness. The authors have identified several baseline features that may help predict the risk of lasting dizziness. Treatment modality does not appear to influence long‐term DHI score. We found a strong association between posttreatment headache and poor dizziness handicap. Future study is needed to further define this relationship.

Object The authors' objective was 2-fold: 1) to compare outcomes of microsurgical resection for vestibular schwannoma (VS) between patients aged 70 years or older and patients younger than 70 years and 2) to test the hypothesis that symptomatic tumors in elderly patients represent a more aggressive variant of VS than those in younger adults, resulting in increased morbidity and a higher rate of recurrence after subtotal resection (STR). Methods A retrospective matched cohort study was conducted. Patients aged 70 years or older who had undergone microsurgical resection of VS were matched to adult patients younger than 70 years; the matching was performed on the basis of surgical approach, completeness of resection, and tumor size. Associations between clinical, radiographic, and surgical data and postoperative outcome were analyzed using conditional logistic regression. Results Twenty patients aged 70 years or older (mean age ± SD 75.9 ± 5.3, range 70–86 years) were identified and matched to a cohort of younger adult patients (mean age ± SD 55.7 ± 13.8, range 25–69 years). The mean tumor size in both groups was approximately 3 cm. Overall, the elderly patients had a poorer preoperative American Society of Anesthesiology physical status score (p = 0.038), were more likely to report imbalance (OR 9.61, p = 0.016), and more commonly exhibited compromised balance and coordination (OR 9.61, p = 0.016) than patients in the younger cohort. There were no differences between the 2 cohorts in perioperative complications (p = 0.26) or facial nerve function (p > 0.5) at any time. The elderly patients were 13 times more likely to have long-term postoperative imbalance (OR 13.00, p = 0.013) than the younger patients. Overall, 9 tumors recurred among 32 patients undergoing STR; 6 of these patients underwent additional interventions (stereotactic radiosurgery in 5 patients and microsurgery in 1) and showed no evidence of tumor progression at the last follow-up. The median growth rate of the recurrent tumor in the 6 elderly patients was 4.8 mm/year (range 2.1–14.9 mm/year) and, in the 3 control patients, 2.2 mm/year (range 1.9–4.0 mm/year). Overall, the mortality data showed a trend toward statistical significance (p = 0.068) with a higher risk of death in the elderly. Conclusions As the number of elderly patients with VS increases, microsurgical resection will remain an important management option for these patients. Despite a poorer preoperative physical status in elderly patients, their morbidity profiles are similar to those in adult patients younger than 70 years. However, elderly patients may require longer convalescence due to prolonged postoperative imbalance. Not surprisingly, overall diminished functional reserve and advanced comorbidities may increase the mortality risk associated with surgical intervention in the elderly patients. Finally, there was a high risk of further tumor growth following STR in the elderly patients (6 [37.5%] of 16), underscoring the need for close postoperative radiological surveillance and consideration of early stereotactic radiosurgery for the tumor remnant following the STR.

Objectives/HypothesisTo describe the presentation and clinical course of cerebellopontine angle (CPA) and internal auditory canal (IAC) lipomas.Study DesignRetrospective cohort study at a tertiary academic referral center.MethodsAll patients presenting with a CPA or IAC mass radiographically consistent with a lipoma on high‐resolution magnetic resonance imaging (MRI) were identified. Data including presenting symptomatology, tumor characteristics, management strategy, and patient course were collected.ResultsBetween 1996 and 2012, 15 patients were diagnosed with a CPA or IAC lipoma at the authors' institution and were included in the analysis. The mean duration of radiological and clinical follow‐up was 3.4 years and 5.1 years, respectively. Eight lesions were confined to the IAC, while seven involved the CPA. The median tumor size at diagnosis was 7.2 mm; one patient demonstrated tumor growth on serial MRI while the remaining subjects did not have radiological progression. The most common presenting symptoms were sensorineural hearing loss (40%) and tinnitus (33%); five patients were diagnosed after incidental discovery on MRI. Fourteen patients were managed with observation, while one subject underwent subtotal resection. None of the observed patients reported worsening symptoms at last follow‐up.ConclusionsWhile rare, lipomas should be included in the differential diagnosis of CPA and IAC lesions. Owing to a generally benign clinical course and high morbidity associated with resection, microsurgery should only be considered in cases of definite tumor enlargement with intractable symptoms from mass effect. Careful radiological evaluation is critical for establishing an accurate diagnosis in order to prevent unnecessary morbidity associated with resection.Level of Evidence2b.

Objectives/HypothesisTo describe the natural history of primary inner ear schwannomas (PIES) and evaluate management outcomes and relationship between PIES location, clinical presentation, and time to diagnosis.Study DesignRetrospective chart review and systematic review of the literature.MethodsVestibular schwannoma confined to or arising from the inner ear were included. PIES classification was based on anatomic subsite(s) involved. Detailed clinical history and outcomes were recorded.ResultsIn a systematic review (1933–2011), including 14 patients from the authors' institution (1999–2009), a total of 72 studies comprising 234 patients were evaluated. Mean follow‐up was 32.8 ± 39.1 months (range, 0–183 months). The cochlea was the most commonly involved subsite (51%). Hearing loss was the most frequent presenting symptom (99%). Vertigo and abnormal balance were more common among tumors involving the vestibular system (P < .01). Average delay between symptom onset and diagnosis was 7.0 ± 8.0 years (median, 5 years; range, 0–40 years). Recent onset hearing loss was more likely to elicit an earlier diagnosis (P = .01). The majority of patients were observed without treatment (53%). Tumor progression was seen in 52% of patients.Conclusions PIES are rare tumors and most commonly involve the cochlea. Tumor location is often associated with clinical presentation and correlates with delay between symptom onset and diagnosis. A watch‐and‐scan approach is the management strategy of choice in the absence of intractable vertigo or extensive tumor growth. The majority of patients report stable or improved symptoms over time, regardless of treatment. Laryngoscope, 123:1957–1966, 2013

Object The goal of vestibular schwannoma surgery is tumor removal and preservation of neural function. Intraoperative facial nerve (FN) monitoring has emerged as the standard of care, but its role in predicting long-term facial function remains a matter of debate. The present report seeks to describe and critically assess the value of applying current at supramaximal levels in an effort to identify patients destined for permanent facial paralysis. Methods Over more than a decade, the protocol for stimulating and assessing the FN during vestibular schwannoma surgery at the authors' institution has consisted of applying pulsed constant-current stimulation at supramaximal levels proximally and distally following tumor resection to generate an amplitude ratio, which subtracted from 100% yields the degree to which the functional integrity of the FN “dropped off” intraoperatively. These data were prospectively collected and additional variables that might impact postoperative FN function were retrospectively reviewed from the medical record. Only patients with anatomically intact FNs and > 12 months of follow-up data were analyzed. Results There were 267 patients available for review. The average posterior fossa tumor diameter was 24 mm and the rate of long-term good (House-Brackmann Grade I–II) FN function was 84%. Univariate logistic regression analysis revealed that prior treatment, neurofibromatosis Type 2 status, tumor size, cerebellopontine angle extension, subjectively thinned FN at the time of operation, minimal stimulation threshold, percent dropoff by supramaximal stimulation (SMS), and postoperative FN function all correlated statistically (p < 0.05) with long-term FN function. When evaluating patients with significant FN weakness at the time of hospital discharge, only the percent dropoff by SMS remained a significant predictor of long-term FN function. However, the positive predictive value of SMS for long-term weakness is low, at 46%. Conclusions In a large cohort of patients, the authors found that interrogating intraoperative FN function with SMS is safe and technically simple. It is useful for predicting which patients will ultimately have good facial function, but is very limited in identifying patients destined for long-term facial weakness. This test may prove helpful in the future in tailoring less than gross-total tumor removal to limit postoperative facial weakness.