Abstract
OBJECTIVE
The majority of research influencing our understanding of vestibular schwannoma (VS) comes from large tertiary referral centers, and as a consequence, is inherently prone to referral bias. The objective of the current study was to characterize tertiary referral center bias in VS research.
STUDY DESIGN
Single-institution retrospective clinical, audiometric, and radiologic review.
SETTING
One tertiary referral center.
PATIENTS
One hundred twenty-three patients with sporadic VS treated at our institution, consisting of a local cohort of all 41 VS patients residing in the same US county as our medical center and a referral cohort of 82 patients from outside counties matched 2:1 based on age, sex, and year of diagnosis.
INTERVENTION
Surgical resection, stereotactic radiosurgery (SRS), observation.
MAIN OUTCOME MEASURES
Clinical manifestations, audiometric and radiologic findings, elected treatment modality.
RESULTS
Local patients had significantly longer follow-up after initial presentation compared to referral patients (median 4.0 vs 2.2 yr; p = 0.007). Referral patients were significantly less likely to have an incidental VS diagnosis (1% vs. 29%; p < 0.001) and presented with worse symptomatology such as hearing loss (80% vs. 54%; p = 0.002), dizziness (72% vs. 44%; p = 0.002), headache (29% vs. 15%; p = 0.075), and other cranial nerve dysfunction (11% vs. 0%; p = 0.029). Audiometrically, referral patients had significantly poorer word recognition scores at presentation (median 70% vs. 90%; p = 0.043). Local patients were more likely to have tumors confined to the IAC compared with referral patients, and had significantly smaller tumors when restricted to this region (median 4.0 vs 6.5 mm; p = 0.005). Referral patients were significantly more likely to undergo definitive management with either radiosurgery or microsurgery following primary evaluation compared to local patients (48% vs. 24%; p = 0.013).
CONCLUSION
These data suggest that the majority of existing literature surrounding VS likely suffers from referral bias, whereby disease characteristics and management decisions are distinct from that of the general VS patient population.
Collapse