Purpose: To determine significance of findings on orbital computed tomography in classifying ocular injuries into Open Globe or Closed Globe
Methods: The study design was prospective observational case series carried between July 2016 and July 2017. About 604patients with suspected foreign body, in our case pellet, were studied. All patients did undergo helical CT direct scanning in axial plane and multi planar reconstruction of coronal and sagittal planes. The images were studied by masked observers, one radiologist and two ophthalmologists.
Results: Statistically significant CT findings for open-globe injury included intraocular foreign body, globe contour change, intraocular air, vitreous haemorrhage, retinal detachment, and Choroidal detachment. The overall sensitivity, specificity and accuracy of CT for determining open-globe injury were89.89%, 92.04% and 90.57% respectively. Positive predictive value ranged from 93.85% to 97.51% and negative predictive value from 76.01% to 84.80%. Positive likelihood ratio and negative likelihood ratio was found to be 11.29(7.05-18.09) and 0.11(0.08-0.15). Furthermore optic nerve sheath haemangioma (n=13), avulsion (n=3), inflammation (n=23), and retro-ocular haemorrhage (n=22) and relative proptosis (n=9) did not correlated significantly to either group. Also intraconal foreign body correlated more to open-globe injury than extraconal foreign body.
Conclusions: CT can be considered initial investigation of choice in such injuries as it is rapid, non-invasive, provides information regarding posterior segment and retro-ocular tissue, accurate localisation of FB and guides management particularly in settings of multisystem injury. We recommend CT as initial investigation of choice in suspected FB injury and occult open-globe injury.