Orbital Computed Tomography Can Predict Poor Visual Outcomes After Eye Trauma

Computed tomography (CT) scans offer valuable prognostic information toward final visual outcomes of patients who have experienced orbital trauma, according to a report published in the Journal of Ophthalmology. Patients with substantial ocular trauma often present to an emergency center in a distressed or semi-conscious condition, with co-existing injuries that prevent sitting up for slit-lamp examination — thus, clinical assessment of the globe is difficult. 

Researchers reviewed records for patients at least 18 years old with ocular trauma who received orbital CT testing at an academic ophthalmology emergency department in China between January 2020 and December 2021. Charts included 200 individuals who were treated for blunt or piercing trauma — chemical burns were excluded. Two experienced radiologist readers were masked to clinical findings. Poor outcome was defined as acuity worse than 20/1000.

The researchers used the binary cross tab model to analyze sensitivity, specificity, and accuracy for prediction of final visual outcome. Resulting data demonstrated good sensitivity of 65.96%, specificity 76.27%, accuracy 69.00%, and high positive predictive value at 86.92% (all 95% CI). However, negative predictive value was low at 48.39%. In other words, there was sufficient positive prediction of poor visual outcome, but absence of CT findings did not exclude the possibility of low vision.

“Abnormal findings in the orbital CT scan of any patient with traumatic eye injury predicts the poor visual outcome with high accuracy, but the lack of these findings does not rule out the potential poor visual outcome,” according to the researchers.

Notably, binary logistic regression analysis indicated 2 significant independent predictors of poor final acuity: disorganized or collapsed globe had a prognostic value of 74.327 (P <.001), and optic canal/nerve injury reflected a value of 31.605 (P <.001), the most significant variables examined.

Hazard ratio (HR) analysis found the 2 above factors (both P <.001), or any 1 of the following findings showed statistical significance for poor outcome: anterior chamber size increase or decrease (P <.001), globe borders haziness (P =.006), hemorrhage in the anterior or posterior chamber (P <.001), or orbital foreign body (P =.024). Regarding the later 4 HR variables, “the use of these factors for the prediction of visual outcomes in clinical practice should be done with other achievable clinical and physical exam findings and caution regarding the limitations associated with the quality of CT scan machines, the correct applied protocols, and the experience of the interpreter.”

Previous studies have revealed that some of the key clinical features significantly impacting final vision after open globe injury include retinal detachment, vitreous hemorrhage, and presenting vision, as well as crystalline lens displacement, and prior keratoplasty. Another report added to this list endophthalmitis, globe rupture, perforating trauma, and relative afferent pupillary defect. Studies of clinical factors did not cover CT results, yet this test has the benefit of being easier to accomplish when a patient is unconscious or unable to work with the doctor, the study authors report.

A limitation of the analysis is that it concentrates on CT findings only — in real-world scenarios, clinicians use both observed clinical and imaging techniques to assess patients. Also, this investigation employed orbital CT thin-cut slicing of 2 mm, compared with typical emergency department trauma imaging of the head using 10 mm cuts. Strengths of the study are a relatively large sample and the addition of novel data regarding CT’s prognostic benefit.