Predictive Factors for Surgical Decision Making in Nonconcussive Traumatic Brain Injury Patients without Immediate Surgery: A Propensity Score Matching Study of Optic Nerve Sheath Diameter, Glasgow Coma Scale, and Rotterdam Computed Tomography Score
Issued Date
2024-01-01
Resource Type
ISSN
18788750
eISSN
18788769
Scopus ID
2-s2.0-85211111761
Pubmed ID
39566738
Journal Title
World Neurosurgery
Rights Holder(s)
SCOPUS
Bibliographic Citation
World Neurosurgery (2024)
Suggested Citation
Wanachiwanawin C., Chatpuwaphat J., Tritrakarn S.o., Chatkaewpaisal A., Tongsai S., Chankaew E., Kaewlai R. Predictive Factors for Surgical Decision Making in Nonconcussive Traumatic Brain Injury Patients without Immediate Surgery: A Propensity Score Matching Study of Optic Nerve Sheath Diameter, Glasgow Coma Scale, and Rotterdam Computed Tomography Score. World Neurosurgery (2024). doi:10.1016/j.wneu.2024.11.008 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/102403
Title
Predictive Factors for Surgical Decision Making in Nonconcussive Traumatic Brain Injury Patients without Immediate Surgery: A Propensity Score Matching Study of Optic Nerve Sheath Diameter, Glasgow Coma Scale, and Rotterdam Computed Tomography Score
Author's Affiliation
Corresponding Author(s)
Other Contributor(s)
Abstract
Objective: To assess the utility of initial computed tomography (CT)–derived optic nerve sheath diameter (ONSD), Glasgow Coma Scale (GCS) score, Rotterdam CT score, and other factors in predicting subsequent surgical intervention in patients with nonconcussive traumatic brain injury (TBI) who did not undergo immediate surgery. Methods: This retrospective study included 251 adult patients (184 men, mean age 50.2 ± 17.7 years) with nonconcussive TBI who underwent initial head CT within 24 hours of trauma and did not undergo immediate surgery. Propensity scores using age, sex, and injury category matched 27 patients who underwent surgery within 48 hours of admission (surgery group) with 224 patients who did not undergo surgery (nonsurgery group). Clinical and CT data were collected from medical and radiological records, including ONSD measurements performed by 3 radiologists and compared between the groups. Results: All patients had nonconcussive TBI with top 3 injuries including subdural, subarachnoid, and parenchymal hemorrhages in 165, 146, and 133 patients, respectively. ONSD differed between the surgery (mean 6.3 ± 0.8 mm) and nonsurgery (mean 6.0 ± 0.7 mm) groups (P = 0.014) only on a univariable analysis. Multivariable analysis revealed initial GCS score of 9–12 and Rotterdam CT score >3 as predictive of surgery with odds ratios of 5.596 (95% confidence interval 2.031–15.421, P < 0.001) and 5.024 (95% confidence interval 1.823–13.847, P = 0.002), respectively. Conclusions: Initial GCS and Rotterdam CT scores could anticipate the need for subsequent surgery among patients with nonconcussive TBI who did not undergo immediate surgery, whereas ONSD could not.