# Prognostic Value of Minnesota Multiphasic Personality Inventory-2 (MMPI-2) Profiles in Predicting Outcomes of Occipital Nerve Stimulation for Refractory Chronic Migraine: A Retrospective Bias-Corrected Multivariable Analysis

## TL;DR

This study found that higher depression scores on a psychological test predict worse outcomes for a migraine treatment called occipital nerve stimulation.

## Contribution

The study demonstrates that MMPI-2 Depression scores independently predict ONS outcomes in refractory chronic migraine patients.

## Key findings

- ONS significantly reduced pain intensity, headache frequency, and drug consumption at 12 months.
- Higher MMPI-2 Depression scores were associated with poorer ONS outcomes.
- The predictive model showed high sensitivity (AUC: 0.96) and good calibration.

## Abstract

Refractory chronic migraine (rCM) is characterized by debilitating headaches that do not respond adequately to conventional medical treatments, leaving patients severely disabled. In these rare cases, central cervical spinal cord stimulation or occipital nerve stimulation (ONS) may offer a potential therapeutic option. However, these techniques are not without risks, lack clear scientific evidence, and impose a significant economic burden. Therefore, it is crucial to identify parameters that can assist physicians in selecting appropriate candidates for implantation. This study aimed to investigate the role of psychological profiles in predicting outcomes for ONS in rCM patients.

We conducted a retrospective analysis on rCM patients treated with ONS at a second-level neuromodulation university facility. These patients were refractory to conventional medical treatments, including onabotulinumtoxin-A injections (monoclonal antibodies targeting CGRP were not yet available). The NRS for migraine intensity, the number of monthly migraine attacks, and drug consumption were assessed at 6-month and 12-month follow-ups post-implant. Psychological profiles were evaluated prior to OCN using the Minnesota Multiphasic Personality Inventory-2 (MMPI-2). A multivariable logistic regression model was developed to predict ONS outcomes, incorporating MMPI-2 as a covariate. The model’s accuracy and performance were assessed through non-parametric bootstrap, calibration, and discrimination analyses.

Twenty-three rCM patients were analysed. ONS was able to significantly reduce the pain intensity, the number of headache attacks per month, and drug consumption compared to pre-treatment levels at both 6- and 12-month follow-ups. The final multivariable logistic model at 12 months showed that the MMPI-2 Depression score was independently and negatively associated with favourable outcomes following ONS (optimism-adjusted OR 0.52, 95% CI 0.21–0.77, p = 0.03). The ROC curve indicated high model sensitivity (AUC: 0.96, 95% CI: 0.88–0.98), and the calibration plot revealed a good fit, with some improvement needed in mid-range predicted probabilities.

ONS significantly reduced pain intensity, headache frequency, and drug use at the 12-month follow-up compared to pre-treatment levels. The MMPI-2 Depression score was an independent predictor of ONS failure, highlighting the importance of comprehensive psychological assessments in patient selection for ONS.