Neurologic Imaging

Visual Loss with a Known Cause: Why Does the ACR Say Imaging Is Usually Not Appropriate?

·
Visual Loss with a Known Cause: Why Does the ACR Say Imaging Is Usually Not Appropriate?

A 72-year-old woman presents to the emergency department with a two-day history of a “curtain” coming down over her left eye, accompanied by a new, severe headache and jaw pain when she chews. An ophthalmology consult confirms a pale, swollen optic disc on fundoscopy. Her erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) labs return markedly elevated. The diagnosis of giant cell arteritis is made, and high-dose steroids are initiated. As the admitting physician, you pause before ordering a stat head and orbit CT, wondering if it’s necessary to rule out other pathology. For this specific scenario—visual loss where the etiology has been identified by exam or labs—the American College of Radiology (ACR) rates nearly all imaging, including CT, MRI, and even simple radiography, as Usually not appropriate. This article explains the clinical reasoning behind this recommendation.

Who Fits This Clinical Scenario?

This guidance applies to a specific and important clinical situation: a patient presenting with visual loss for whom a confident diagnosis has already been established through non-radiologic means. The key is that the “why” behind the vision loss is no longer a mystery.

Inclusion criteria for this workflow:

  • A definitive finding on a comprehensive ophthalmologic examination (e.g., slit-lamp, tonometry, dilated fundoscopy) that explains the visual symptoms. Examples include retinal detachment, central retinal artery occlusion (CRAO), or acute angle-closure glaucoma.
  • Characteristic laboratory results that confirm a systemic diagnosis responsible for the visual loss, such as highly elevated inflammatory markers in a patient with classic symptoms of giant cell arteritis (GCA).
  • Findings on ancillary ophthalmic testing like optical coherence tomography (OCT) or fluorescein angiography that pinpoint the pathology.

Exclusion criteria (these patients belong to different workflows):

  • Suspected Optic Neuritis: Patients with painful vision loss, often with a relative afferent pupillary defect, but a normal-appearing optic disc on initial exam. This presentation requires a different imaging pathway, typically starting with MRI.
  • Unexplained Visual Field Defects: Patients with visual loss patterns suggesting a lesion at or behind the optic chiasm (e.g., bitemporal hemianopsia). This points toward a chiasmal or post-chiasm workup.
  • Orbital Trauma: Any patient with a traumatic mechanism for their visual defect requires an initial imaging workup focused on identifying fractures, hemorrhage, or foreign bodies.

Correctly categorizing the patient is the critical first step. If the cause is truly known, imaging is unlikely to add value. If there is any diagnostic uncertainty, this guidance does not apply.

What Diagnoses Are You Working Up in This Scenario?

In this context, the “workup” is already complete, and imaging is being considered as a confirmatory or exclusionary step. The diagnoses that fall into this category are those readily identified at the bedside or with basic labs.

A common example is Giant Cell Arteritis (GCA). In an older adult with new-onset headache, jaw claudication, and vision loss, highly elevated ESR and CRP levels are strongly suggestive. The vision loss is typically due to arteritic anterior ischemic optic neuropathy (A-AION), a diagnosis made clinically. Treatment with high-dose corticosteroids should not be delayed for imaging.

Another major category is Retinal Vascular Occlusions. A central retinal artery occlusion (CRAO) presents with sudden, profound, painless vision loss and has a classic “cherry-red spot” on fundoscopy. A central retinal vein occlusion (CRVO) has a different but equally characteristic “blood and thunder” appearance. In both cases, the diagnosis is made by looking at the retina, not with a CT or MRI.

Retinal Detachment is also diagnosed on direct visualization. The patient’s history of floaters, flashes of light (photopsia), and a curtain-like visual field defect, confirmed by a fundoscopic exam showing the detached retinal tissue, is definitive. The immediate next step is surgical consultation, not cross-sectional imaging.

Other etiologies include Acute Angle-Closure Glaucoma, diagnosed with tonometry (high intraocular pressure) and gonioscopy, and complications of severe systemic disease like a vitreous hemorrhage in a patient with proliferative diabetic retinopathy, which is also seen on exam.

Why Is Advanced Imaging Usually Not Appropriate for This Presentation?

When a clear diagnosis for visual loss has been made via ophthalmologic exam or lab tests, the ACR Appropriateness Criteria rate all major imaging modalities as Usually not appropriate. This includes MRI of the orbits, CT of the head, and even basic orbital radiography. The core rationale is that these studies provide no meaningful benefit, do not change immediate management, and can introduce unnecessary risks and delays.

The diagnostic question has been answered. Ordering a scan at this stage is a low-yield endeavor. For a patient with a confirmed retinal detachment, an MRI of the orbits adds no useful information for the retinal surgeon and only delays definitive care. Similarly, for a patient with GCA, a CT scan will not visualize the inflammation in the small temporal or ophthalmic arteries and should not postpone the urgent administration of steroids.

Let’s examine why specific, powerful alternatives are still not recommended here:

  • MRI orbits without and with IV contrast (Radiation: O 0 mSv): While this is the most sensitive test for optic nerve and intraconal pathology, it is rated Usually not appropriate. In the case of a CRAO or GCA, the pathology is at a microscopic or vascular level that is not resolved by standard MRI. The test is expensive, time-consuming, and offers no advantage over the clinical diagnosis, potentially delaying time-sensitive treatments.
  • CT orbits without IV contrast (Radiation: ☢☢☢ 1-10 mSv): This modality is also rated Usually not appropriate. It exposes the patient to ionizing radiation without being able to visualize the retina, optic nerve ischemia, or vascular inflammation. It is excellent for bone and acute hemorrhage but is the wrong tool for the pathologies in this scenario.

The fundamental principle is clinical parsimony. Once the cause is identified, the focus must shift from diagnosis to treatment. Imaging in this context risks delaying care, increasing costs, and potentially uncovering incidental findings that lead to a cascade of further, unnecessary investigation.

What’s Next After the Diagnosis? Downstream Workflow

With imaging off the table, the workflow pivots immediately to pathology-specific management. The next steps are dictated by the diagnosis established by the ophthalmologist or lab results.

  • If the diagnosis is Giant Cell Arteritis: The immediate next step is administering high-dose systemic corticosteroids to prevent further ischemic events, including vision loss in the contralateral eye. This should be followed by an urgent rheumatology consultation and arrangements for a temporal artery biopsy to confirm the diagnosis histologically.
  • If the diagnosis is Central Retinal Artery Occlusion: Management is time-sensitive and focused on attempts to dislodge the embolus, such as ocular massage or anterior chamber paracentesis, though efficacy is debated. The crucial downstream workflow is a systemic workup for the embolic source, which includes carotid artery imaging (ultrasound or CTA/MRA), an echocardiogram, and long-term cardiac monitoring. This is a stroke workup, not an orbital imaging workup.
  • If the diagnosis is Retinal Detachment: This is a surgical emergency. The only appropriate next step is an immediate referral to a vitreoretinal surgeon for evaluation and repair.
  • If the clinical picture changes or is incongruent: If a patient develops new neurologic symptoms that are not explained by the ocular diagnosis (e.g., hemiparesis, cranial nerve palsies), the clinical scenario has changed. At that point, the patient no longer fits this specific guideline. The presentation may have evolved into a different scenario, such as nonischemic visual loss with post-chiasm symptoms, which would trigger a completely different imaging workup (typically MRI head).

Pitfalls to Avoid (and When to Get Help)

Navigating this scenario requires confidence in the initial non-imaging diagnosis. Here are common pitfalls:

  • Reflexive Imaging: The most common error is ordering a CT or MRI “just to be safe” despite a clear diagnosis. This delays definitive treatment, adds unnecessary cost, and exposes the patient to risk.
  • Misclassifying the Patient: Be certain the diagnosis is secure. If the ophthalmologic exam is equivocal or the lab results are borderline, the patient may not truly fit this “etiology identified” scenario and may require imaging to resolve the diagnostic uncertainty.
  • Ignoring New Neurologic Deficits: Do not attribute all new symptoms to the known ocular diagnosis. A patient with a CRVO who subsequently develops aphasia needs an urgent stroke workup; the two events may be related systemically but require separate diagnostic pathways.

If the clinical findings and the patient’s symptoms seem incongruent, or if new, unexplained neurologic signs emerge, escalate immediately. This often involves a neurology consultation and a shift in the diagnostic paradigm to one that warrants neuroimaging.

Related ACR Topics and Tools

This article covers a single, specific clinical scenario. For a comprehensive overview of all related presentations, from orbital trauma to suspected optic neuritis, please consult the parent topic guide. For additional resources on applying appropriateness criteria and understanding imaging protocols, the following tools are available.

Frequently Asked Questions

Are there any exceptions where an identified etiology still requires imaging?

Rarely. One example could be a known orbital tumor, diagnosed via biopsy, that requires imaging for surgical planning or to monitor for growth. Another might be a patient with a CRAO where a carotid dissection is suspected as the embolic source, which would require CTA or MRA of the head and neck. However, in these cases, the imaging is not for diagnosing the cause of vision loss, but for managing the underlying, already-identified condition.

My ophthalmology colleague confirmed the diagnosis but still asked for an MRI. Why?

This can happen if there are atypical features or the findings are not fully explained by the diagnosis. For example, if a patient has optic disc edema that seems out of proportion to the findings of a central retinal vein occlusion, the ophthalmologist might request an MRI to rule out a compressive optic neuropathy. This indicates the diagnosis is not considered fully secure, and the patient technically falls into a different clinical scenario.

What if the patient’s vision loss is functional or non-organic?

Functional (or non-organic) vision loss is a diagnosis of exclusion after a thorough workup, including a normal ophthalmologic exam. In that case, the etiology is not truly ‘identified’ in the same way as a retinal detachment. A clinician might consider neuroimaging (typically MRI) to confidently rule out subtle organic pathology before making a diagnosis of functional vision loss. This patient would not fit the scenario described in this article.

Does this ‘no imaging’ recommendation apply to both adults and children?

Yes, the principle applies regardless of age. If a definitive diagnosis for the visual loss is established in a pediatric patient via exam (e.g., congenital glaucoma, traumatic hyphema), then further imaging is generally not warranted unless it’s needed for management of that specific condition (e.g., CT for a complex orbital fracture). The goal of avoiding unnecessary radiation is even more critical in children.

If a patient has a CRAO, isn’t a brain MRI needed to look for a stroke?

A central retinal artery occlusion is often considered a form of stroke. A brain MRI with diffusion-weighted imaging (DWI) is highly sensitive for detecting concurrent cerebral ischemia, which occurs in a significant number of CRAO patients. However, this MRI is part of the systemic stroke workup to assess risk and guide secondary prevention; it is not an ‘orbital’ scan to find the cause of the vision loss itself, which is already known. The decision to order a brain MRI in this context is part of the stroke management pathway.

Reviewed by Pouyan Golshani, MD, Interventional Radiologist — May 29, 2026