To image or not to image? A cost‐effectiveness analysis of MRI for patients with asymmetric sensorineural hearing loss

Objectives/Hypothesis Our objective was to perform an economic analysis evaluating whether ordering a magnetic resonance imaging (MRI) is a cost-effective practice in the workup of undifferentiated asymmetric sensorineural hearing loss (ASNHL). Use of T1 gadolinium-weighted MR (GdT1W) and T2 weighted MR without contrast (T2MR) was each examined. Methods The incremental cost-effectiveness ratio (ICER) of MR imaging among ASNHL patients was evaluated using a decision tree. We calculated what the probability of having a cerebellopontine angle/internal auditory canal lesion would have to be to make MR more cost-effective than observation. The decision pathways included observation, GdT1W, and T2MR. The probability of detecting a lesion and associated costs were employed in constructing our tree. Missing a mass in the observation branch was considered to have an effectiveness of 0. The costs and probabilities were extracted from previously published studies. Results The ICERs of pursuing GdT1W and T2MR were $27,660 and $15,943, respectively, both below the widely accepted willingness to pay (WTP) thresholds of $30,000 and $50,000. Probabilistic sensitivity analysis with Monte Carlo simulations for GdT1W showed that it is more cost-effective than no imaging, with 54.4% and 83.5% certainty at $30,000 and $50,000 WTP thresholds. Probabilistic sensitivity analysis with Monte Carlo simulations for T2MR showed that it is more cost-effective than no imaging, with 75.2% and 92.6% certainty at $30,000 and $50,000 WTP thresholds. Conclusion This economic evaluation strongly supports pursuing MRI in patients with documented ASNHL as a cost-effective strategy. Both GdT1W and T2MR are more cost-effective than observation. Furthermore, noncontrast T2 imaging may be the more cost-effective modality of these two techniques. Level of Evidence 2c. Laryngoscope, 2016