• Skip to main content
  • Skip to header right navigation
  • Skip to site footer

786-300-3183 | [email protected]

  • Twitter
  • Facebook
Florida Society of Anesthesiologists

Florida Society of Anesthesiologists

  • About FSA
    • FSA Leadership
      • FSA Past Presidents
      • Distinguished Service Award Past Recipients
      • Recipients of the FSA Presidential Engagement Award
    • FSA Staff
    • FSA NEWS
    • Calendar of Events
    • Contact FSA
    • FSA Charter & Bylaws
    • FSA Speakers Bureau
  • FSA Annual Meeting
    • 2025 Annual Meeting
    • 2024 Annual Meeting Recap
    • Call For Abstracts
    • Past Posters
      • 2025 FSA Podium and Poster Abstracts
      • 2024 FSA Podium and Poster Abstracts
      • 2023 FSA Podium and Poster Abstracts
      • 2022 FSA Podium and Poster Abstracts
      • 2021 FSA Posters
      • 2020 FSA Posters
      • 2019 FSA Posters
      • 2018 FSA Posters
    • Past Meetings
      • 2023 Meeting Recap
      • 2022 Annual Meeting Recap
      • 2019 Annual Meeting Recap
      • 2018 Annual Meeting Recap
  • FSAPAC
    • Donate to the FSAPAC
  • Member Login
  • Member Portal
  • Become a Member
    • FSA Membership Renewal
    • Join the Florida Society of Anesthesiologists (FSA)

2019 FSA Posters

2019 FSA Posters

P057: PREDICTIVE RISK MODELING FOR CERVICOGENIC HEADACHE FOLLOWING CERVICAL RADIOFREQUENCY NEUROTOMY
Mollie K Mansfield, BS, Joseph E LaGrew, MD, Terrie Vasilopoulos, PhD, Sanjeev Kumar, MD; University of Florida Department of Anesthesia

Introduction: Percutaneous radiofrequency neurotomy (PRN) has been shown to be a valid treatment for zygapophyseal pain,1 but treatment of one source of neck pain could lead to unmasking or even perceived exacerbation of nociception from other pain generators.2 Previous work has demonstrated the phenomenon of new onset of cervicogenic headache (CGH) following lower cervical PRN.3 In this retrospective cohort study, risk factors for CGH following radiofrequency neurotomy were defined.

Methods: This single center study included a review of 326 patients receiving cervical PRN at the University of Florida Pain Management Center from 2014 to 2016, as previously described.3 Risk factors with p-values < 0.05 from univariate analysis were included in multivariate analysis with best subset approach and AIC validation.4[V1]  Weights for the risk score were calculated by first multiple the regression coefficient by 10, dividing by lowest value, then rounding to nearest integers. Receiver operator characters (ROC) curve analysis was used to determine a cutoff value for the risk score that maximized sensitivity and specificity.

Results: Eight potential risk factors were identified from univariate analysis of with Five factors emerged as significant independent predictors of CGH: worst pre-procedure pain level of 9 or 10 (p= 0.047), greater than or equal to 3 total treatments, (p = 0.017), comorbid diabetes (p = 0.001), absence of shoulder pain on presentation (p = 0.033), pain alleviated by exercise (p = 0.037), and pain not relieved by rest (p = 0.013). The ROC curve for the scale is depicted in Figure 1, with an AUC = 0.901 (95%CI 0.786 – 0.987). Weighted risk factors with a maximum score of 9 was found to have optimal sensitivity, specificity and negative based on scores greater than 4 or less than/equal to 4. 

Conclusion: This analyses yielded a predictive model to better define the risk of CGH in patients undergoing lower cervical PRN. Further work is needed to prospectively validate the use of this of this model in a large cohort.

Copyright © 2025 · Florida Society of Anesthesiologists · All Rights Reserved