Can a computed tomography (CT) scan without contrast detect bleeding in the eye?

Can CT Without Contrast Detect Bleeding in the Eye?

Yes, CT without contrast can detect intraocular bleeding, but its sensitivity is limited compared to other modalities, and it should not be the first-line diagnostic approach for suspected eye hemorrhage.

Diagnostic Performance of Non-Contrast CT for Intraocular Bleeding

Standard head CT without contrast has poor sensitivity for detecting intraocular hemorrhages, with a sensitivity of only 60% for vitreous/preretinal hemorrhages and 32% for any ocular bleeding. 1 This means that CT will miss approximately 40-68% of intraocular bleeds, making it an unreliable screening tool when eye bleeding is the primary concern.

What CT Can and Cannot Detect

Non-contrast CT performs differently depending on the type and location of intraocular bleeding:

• Vitreous and preretinal hemorrhages: Detected in approximately 60% of cases on standard head CT 1
• Retinal hemorrhages: Very poorly detected, with sensitivity of only 32% 1
• Larger hemorrhages: More likely to be visible, but small bleeds are frequently missed 2

The attenuation characteristics of acute blood on non-contrast CT can help identify hemorrhage when present, with high attenuation indicating acute bleeding. 3 However, the small size of the globe and limited spatial resolution of standard head CT protocols significantly limit detection capability.

Superior Alternative Imaging Modalities

Ocular B-Mode Ultrasound (Recommended First-Line)

Ocular sonography is the superior bedside diagnostic tool for intraocular bleeding, with 100% sensitivity and specificity for vitreous and preretinal hemorrhages. 1 This makes it far more reliable than CT for detecting eye bleeding. Ultrasound should be performed by an experienced examiner and can be done at the bedside in the emergency department. 4, 5

MRI of the Orbits

MRI provides excellent soft-tissue resolution for evaluating intraocular pathology and is the preferred modality for detailed assessment of globe and orbital structures. 3 MRI of the orbits without and with contrast is recommended when comprehensive evaluation of intraocular masses or optic nerve pathology is needed. 3

However, standard T1-weighted MRI has very poor sensitivity (only 11.1%) for detecting intraocular glass fragments and associated hemorrhage. 2

Clinical Algorithm for Suspected Intraocular Bleeding

Step 1: Clinical Examination

• Perform funduscopy as the gold standard for detecting retinal, preretinal, and vitreous hemorrhages 1
• Check visual acuity, pupil examination, external examination, and slit lamp examination 5

Step 2: Initial Imaging Selection

• If funduscopy is limited or inconclusive: Obtain ocular B-mode ultrasound as the first imaging study 1
• If trauma with possible foreign body: Obtain helical CT (with or without contrast) as it has 57% detection rate for intraocular glass 2
• If comprehensive orbital evaluation needed: Obtain MRI orbits without and with contrast 3

Step 3: When Standard Head CT May Incidentally Detect Eye Bleeding

• If head CT is being performed for other indications (e.g., suspected intracranial hemorrhage), it may incidentally detect large intraocular hemorrhages 1
• Do not rely on negative head CT to exclude intraocular bleeding given its poor sensitivity 1

Critical Pitfalls to Avoid

Never use standard head CT as the primary screening tool for suspected intraocular hemorrhage. 1 The high false-negative rate (40-68%) means significant bleeds will be missed, potentially delaying appropriate ophthalmologic intervention.

Do not assume that CT with contrast offers significant advantage over non-contrast CT for detecting intraocular bleeding. 3 The guidelines do not support routine use of contrast-enhanced CT specifically for intraocular hemorrhage detection, and contrast is typically reserved for evaluating orbital masses or vascular lesions. 3

Recognize that hyphema (blood in the anterior chamber) does not significantly affect CT detection rates. 2 However, the presence of intraocular blood can complicate detection of other pathology like foreign bodies.

Special Considerations

Terson's Syndrome (SAH-Related Eye Bleeding)

In patients with subarachnoid hemorrhage, ocular sonography should be considered as a routine bedside screening tool, as it is superior to standard head CT for detecting SAH-related intraocular hemorrhages. 1

Foreign Body Detection

Helical CT is the most sensitive imaging modality for detecting intraocular glass foreign bodies (57% detection rate), significantly better than axial CT (41%) or MRI (11%). 2 Detection improves with larger fragment size (96% for 1.5mm fragments vs. 48% for 0.5mm fragments) and varies by glass type (90% for green beer bottle glass vs. 43% for spectacle glass). 2