Obesity Is Not Associated With Postoperative Complications After Vestibular Schwannoma Surgery in a Large Single Institution Series

A multi-institutional machine learning algorithm for prognosticating facial nerve injury following microsurgical resection of vestibular schwannoma

Vestibular schwannomas (VS) are the most common tumor of the skull base with available treatment options that carry a risk of iatrogenic injury to the facial nerve, which can significantly impact patients' quality of life. As facial nerve outcomes remain challenging to prognosticate, we endeavored to utilize machine learning to decipher predictive factors relevant to facial nerve outcomes following microsurgical resection of VS. A database of patient-, tumor- and surgery-specific features was constructed via retrospective chart review of 242 consecutive patients who underwent microsurgical resection of VS over a 7-year study period. This database was then used to train non-linear supervised machine learning classifiers to predict facial nerve preservation, defined as House-Brackmann (HB) I vs. facial nerve injury, defined as HB II-VI, as determined at 6-month outpatient follow-up. A random forest algorithm demonstrated 90.5% accuracy, 90% sensitivity and 90% specificity in facial nerve injury prognostication. A random variable (rv) was generated by randomly sampling a Gaussian distribution and used as a benchmark to compare the predictiveness of other features. This analysis revealed age, body mass index (BMI), case length and the tumor dimension representing tumor growth towards the brainstem as prognosticators of facial nerve injury. When validated via prospective assessment of facial nerve injury risk, this model demonstrated 84% accuracy. Here, we describe the development of a machine learning algorithm to predict the likelihood of facial nerve injury following microsurgical resection of VS. In addition to serving as a clinically applicable tool, this highlights the potential of machine learning to reveal non-linear relationships between variables which may have clinical value in prognostication of outcomes for high-risk surgical procedures.

Impact of Obesity on Postoperative Complications after Lateral Skull Base Surgery: A Systematic Review

INTRODUCTION

The relationship between obesity and complications after lateral skull base tumor resection is not clear. There is conflicting evidence regarding the incidence of postoperative complications in this patient population. The purpose of this study is to examine the relationship between obesity and outcomes following lateral skull base tumor resection.

DATA SOURCES

Data were extracted from PubMed, Embase, CINAHL, and Cochrane CENTRAL.

METHODS

Included studies assessed the relationship between obesity and outcomes following lateral skull base tumor removal. Studies with ≤5 patients, pediatric patients, duplicate patient populations, or insufficient data were excluded. Two independent investigators reviewed each study for inclusion. A third reviewer served as a tie-breaker for any conflicts. Extracted data includes patient demographics, tumor pathology, surgical approach, and postoperative outcomes including incidence of cerebrospinal fluid (CSF) leak and other postoperative complications, length of stay (LOS), and readmission and reoperation rates. Descriptive statistics were used to compare postoperative outcomes for obese and nonobese controls.

RESULTS

14 studies met final inclusion criteria. Nine studies evaluated the relationship between obesity and CSF leaks. Four studies found a significant increase in postoperative CSF leak in obese patients compared to nonobese controls. The remaining studies trended toward an increased incidence of CSF leak in the obese population but did not reach statistical significance. One out of seven studies found that obesity increased postoperative LOS, and one out of five studies found that obesity increased reoperation rates following tumor resection.

CONCLUSIONS

Based on the results, obesity does not appear to increase LOS, readmission, or reoperation rates after lateral skull base tumor resection. The relationship between obesity and postoperative CSF leak, however, warrants further analysis.

Elevated Body Mass Index Associated with Cerebrospinal Fluid Leak after Lateral Skull Base Surgery: A Systematic Review and Meta-analysis

OBJECTIVE

To determine if body mass index (BMI) increases the risk of cerebrospinal fluid (CSF) leak after lateral skull base surgery.

DATA SOURCES

CINAHL, PubMed, and Scopus were searched from January 2010 to September 2022 for articles published in English.

STUDY SELECTION

Articles that reported BMI or obesity with and without CSF leaks after lateral skull base surgery were included.

DATA EXTRACTION

Two reviewers (F.G.D. and B.K.W.) independently performed study screening, data extraction, and risk of bias assessment.

DATA SYNTHESIS

A total of 11 studies and 9,132 patients met inclusion criteria. Meta-analysis of mean difference (MD), odds ratio (OR), proportions, and risk ratio (RR) were calculated using RevMan 5.4 and MedCalc 20.110. BMI for patients with CSF leak after lateral skull base surgery (29.39 kg/m2, 95% confidence interval [CI] = 27.75 to 31.04) was significantly greater than BMI for patients without CSF leak after lateral skull base surgery (27.09 kg/m2, 95% CI = 26.16 to 28.01) with an MD of 2.21 kg/m2 (95% CI = 1.09 to 3.34, p = 0.0001). The proportion of patients with BMI ≥ 30 kg/m2 that had a CSF leak was 12.7%, and the proportion of patients with BMI < 30 kg/m2 (control) that had a CSF leak was 7.9%. The OR for CSF leak after lateral skull base surgery in patients with BMI ≥ 30 kg/m2 was 1.94 (95% CI = 1.40 to 2.68, p < 0.0001), and the RR was 1.82 (95% CI = 1.36 to 2.43, p < 0.0001).

CONCLUSION

Elevated BMI increases the risk of CSF leak after lateral skull base surgery.

LEVEL OF EVIDENCE

IIa.

Progression of Hearing Loss in Observed Non-Growing Vestibular Schwannoma

OBJECTIVE

To assess hearing outcomes in observed vestibular schwannoma (VS) with focus on non-growing tumors.

STUDY DESIGN

Retrospective review.

SETTING

Two tertiary neurotology centers.

PATIENTS AND INTERVENTIONS

Patients with sporadic VS undergoing at least 3 years' observation.

MAIN OUTCOME MEASURES

Changes in pure tone averages (PTA) and word recognition scores (WRS) normalized to the contralateral ear.

RESULTS

During the study period, 39 of 105 included patients (37.1%) had tumor growth. Patients with tumor growth had a mean normalized increase in PTA of 8.0 dB HL (p = 0.008) corresponding to a normalized average worsening of their PTA of 1.8 dB per year. Patients with non-growing tumors less than 5 mm in maximal dimension did not have significant ongoing normalized hearing loss (p > 0.05). Patients with non-growing tumors more than or equal to 5 mm had a mean normalized increase in PTA of 7.4 dB HL (p = 0.001) corresponding to an average of 2.0 dB HL per year, which was similar to the loss observed in growing tumors regardless of size (p > 0.05). Normalized decline in PTA of at least 5 dB HL was seen in 72% of patients with growing tumors, 53% of patients with nongrowing tumors more than or equal to 5 mm, and 38% of patients with non-growing tumors less than 5 mm.

CONCLUSIONS

A long-term analysis of hearing outcomes in observed vestibular schwannoma is presented. With observation, VS more than or equal to 5 mm is associated with continued hearing loss even without tumor growth, while non-growing tumors less than 5 mm are not associated with continuing hearing loss. These data inform expectations for observed VS for providers and patients.

Automated objective surgical planning for lateral skull base tumors

PURPOSE

Surgical removal of pathology at the lateral skull base is challenging because of the proximity of critical anatomical structures which can lead to significant morbidity when damaged or traversed. Pre-operative computed surgical approach planning has the potential to aid in selection of the optimal approach to remove pathology and minimize complications.

METHODS

We propose an automated surgical approach planning algorithm to derive the optimal approach to vestibular schwannomas in the internal auditory canal for hearing preservation surgery. The algorithm selects between the middle cranial fossa and retrosigmoid approach by utilizing a unique segmentation of each patient's anatomy and a cost function to minimize potential surgical morbidity.

RESULTS

Patients who underwent hearing preservation surgery for vestibular schwannoma resection (n = 9) were included in the cohort. Middle cranial fossa surgery was performed in 5 patients, and retrosigmoid surgery was performed in 4. The algorithm favored the performed surgical approach in 6 of 9 patients.

CONCLUSION

We developed a method for computing morbidity costs of surgical paths to objectively analyze surgical approaches at the lateral skull base. Computed pre-operative planning may assist in surgical decision making, trainee education, and improving clinical outcomes.

Investigating Predictors of Increased Length of Stay After Resection of Vestibular Schwannoma Using Machine Learning

OBJECTIVE

To evaluate the predictors of prolonged length of stay (LOS) after vestibular schwannoma resection.

STUDY DESIGN

Retrospective chart review.

SETTING

Tertiary referral center.

PATIENTS

Patients who underwent vestibular schwannoma resection between 2008 and 2019.

INTERVENTIONS

Variables of interest included age, body mass index, comorbidities, symptoms, previous intervention, microsurgical approach, extent of resection, operative time, preoperative tumor volume, and postoperative complications. Predictive modeling was done through multivariable linear regression and random forest models with 80% of patients used for model training and the remaining 20% used for performance testing.

MAIN OUTCOME MEASURES

LOS was evaluated as the number of days from surgery to discharge.

RESULTS

Four hundred one cases from 2008 to 2019 were included with a mean LOS of 3.0 (IQR = 3.0-4.0). Postoperatively, 14 (3.5%) of patients had LOS greater than two standard deviations from the mean (11 days). In a multivariate linear regression model (adjusted R2 = 0.22; p < 0.001), preoperative tumor volume (p < 0.001), coronary artery disease (p = 0.002), hypertension (p = 0.029), and any major complication (p < 0.001) were associated with increased LOS (by 0.12, 3.79, 0.87, and 3.20 days respectively). A machine learning analysis using a random forest identified several potential nonlinear relationships between LOS and preoperative tumor dimensions (length, volume) and operative time that were not captured on regression. The random forest model had lower prediction error compared to the regression model (RMSE 5.67 vs. 44.59).

CONCLUSIONS

Tumor volume, coronary artery disease, hypertension, and major complications impact LOS. Machine learning methods may identify nonlinear relationships worthy of targeted clinical investigation and allow for more accurate patient counseling.

Clinical Predictors of Facial Nerve Outcomes After Surgical Resection of Vestibular Schwannoma

OBJECTIVE

To identify clinical predictors of facial nerve (FN) outcomes after microsurgical resection of vestibular schwannoma (VS).

STUDY DESIGN

Prospective cohort study.

SETTING

Academic medical center.

METHODS

Consecutive patients undergoing VS resection from November 2017 to October 2019 were included. FN function was evaluated with the House-Brackmann (HB) scale and stratified into good (HB I-II) and poor (HB III-VI) function. Analyses included descriptive statistics, correlation, and logistic regression.

RESULTS

Of 256 patients who met criteria (mean age, 47.7 years; 62.5% female), 227 (88.7%) achieved good FN function postoperatively and 238 (93.0%) at latest follow-up (mean, 154.8 days). Operative approaches consisted of translabyrinthine (50.8%), retrosigmoid (25.0%), and middle fossa craniotomies (24.2%). Extent of resection was decided intraoperatively, and gross or near total resection was accomplished in 237 (92.6%) cases. Postoperative HB grade correlated with latest HB grade (0.615, P < .001). Factors associated with good postoperative FN function included small tumor size (≤15 mm; odds ratio [OR], 2.425; P = .042), gross or near total resection (OR, 3.170; P = .041), and ≥100-µV intraoperative FN electromyographic response to a 0.05-mA stimulus (OR, 22.242; P < .001). Factors associated with good FN function at latest follow-up included gross total resection (OR, 7.764; P = .003) and ≥100-µV FN electromyographic response (OR, 8.518; P < .001), accounting for surgical approach and tumor size.

CONCLUSION

Microsurgical resection of VS can be accomplished with excellent FN outcomes. Gross total resection and ≥100-µV intraoperative FN electromyographic response predicted excellent FN outcomes. Immediate postoperative FN function is a prognosticator of long-term FN function.

Early Adverse Events Following Transcervical Hypopharyngeal Diverticulum Surgery

Objective:

To describe rates of short-term morbidity following transcervical surgical treatment of hypopharyngeal diverticulum (HD) and analyze predictors of adverse events.

Methods:

The 2005 to 2017 National Surgical Quality Improvement Program database was queried to identify patients diagnosed with HD undergoing open transcervical diverticulectomy.

Results:

A total of 597 patients with a mean age of 71.4 ± 12.3 years were included. Thirty-day adverse events were experienced by 63 (10.6%) subjects, including 6.5% unplanned reoperations, 4.2% surgical complications, 4.4% medical complications, 2.7% readmissions, and 0.7% deaths. Medical complications notably included pneumonia (2.0%), reintubation (1.2%), sepsis (1.2%), intubation >48 hours (0.5%), urinary tract infection (0.5%), or deep vein thrombosis (0.5%), while surgical complications included organ/space infection (2.0%) and superficial (1.3%) or deep (1.0%) surgical site infection. Gender, race, functional status, diabetes, dyspnea, hypertension, steroid use, and recent weight loss were not associated with adverse events. Length of operation and hospitalization were both higher among those with adverse events (127.4 ± 107.9 vs 95.7 ± 59.8 minutes, P = 0.027, and 7.4 ± 7.4 vs 2.8 ± 3.6 days, P < 0.001). On multivariable logistic regression, high American Society of Anesthesiologists (ASA) class (OR = 2.02, P = 0.017), smoking (OR = 2.10, P = 0.044), and operation time (OR = 1.01; P = 0.005) were independent predictors of adverse events. Obesity was not associated with length of stay, readmission/reoperation, or complications. However, increased age was associated with shorter operations ( P = 0.020), higher length of hospitalization ( P < 0.001), and higher mortality ( P = 0.027) and readmission rates ( P = 0.023).

Conclusion:

Understanding clinical factors associated with complications following open surgery for HD such as ASA score, smoking status, length of operation, and age can help optimize surgical outcomes for at-risk patients.

The association of age, body mass index, and frailty with vestibular schwannoma surgical morbidity

OBJECTIVE

To evaluate whether increased body mass index (BMI), age, or frailty influence vestibular schwannoma (VS) short-term surgical morbidity.

METHODS

The 2005-2017 National Surgical Quality Improvement Program database was queried for patients with VS undergoing surgical resection. Age was stratified according to age <50, 50-64, and ≥65, while BMI was stratified based on a threshold of 30. Frailty score (0-5) was indicated based on functional status, diabetes, chronic obstructive pulmonary disease, congestive heart failure, and hypertension.

RESULTS

A total of 1405 patients were included consisting of 56.7 % females with a mean age of 50.7 ± 13.8 years and mean BMI of 29.4 ± 6.6. Patients <50 (n = 604), 50-64 (n = 578), and ≥65 (n = 223), had different duration of surgery (428 ± 173 vs. 392 ± 149 vs. 387 ± 154 min; p < 0.001) and 30-day mortality rates (0.7 % vs. 0% vs. 1.8 %; p = 0.01). However, post-operative length of stay (LOS) (p = 0.16), readmission (p = 0.08), reoperation (p = 0.54), and complication rates were similar. Post-operative myocardial infarction (p = 0.03) and wound infection (p = 0.02) were more commonly observed in the obese cohort (BMI≥30) but readmission (p = 0.18), reoperation (p = 0.44), and complication rates were similar to those with BMI<30. Severely obese patients (BMI≥35) also had higher rates of deep vein thrombosis (p = 0.004). Frailty score 0 (n=921), 1 (n=375), and 2-4 (n=109) was associated with LOS (4.7±3.5 vs. 5.3 ± 4.1 vs. 6.7 ± 6.6 days, p < 0.001) and prolonged intubation rates (1.0 % vs. 2.4 % vs. 3.7 %; p = 0.03).

CONCLUSIONS

Increased age, BMI, and frailty among VS patients were associated with different post-operative complication rates, operation time, or LOS. Knowledge of these can optimize care for at-risk patients.