Transient Monocular Visual Loss Due to Primary Vasospasm: A Case Study
JACQUELINE M.S. WINTERKORN, MD, PhD
A 42-year-old man experienced repeated attacks of transient monocular visual loss (TMVL) in his left eye for several months. The episodes occurred 3 to 6 times per week and lasted about 5 minutes. Each episode was accompanied by a sense of awareness of the left eye, but was without pain or headache. Vision was lost over about a minute during which a gray cloud came over the lower hemifield and then engulfed the entire visual field of the left eye. He believed he could tell light from dark through the cloud. There were no scintillations or zigzag lines. A left relative afferent pupillary defect could be elicited while the gray cloud continued to block vision. The cloud then broke up and cleared over a minute. The attacks occurred at all times of day, in various postures, and even when he was just sitting still. Between attacks he felt well and his vision returned to normal. He had an occasional tension headache but no family or personal history of migraine. He had neither hypertension nor diabetes. His only medication was dipyridamole with aspirin 25 mg (Aggrenox, Boehringer Ingelheim) prescribed by the first doctor to whom he had described the symptoms.
DIAGNOSIS AND TREATMENT
The first step in developing a differential diagnosis of transient visual loss is to ascertain whether the defect occurred in 1 eye or both eyes. The patient with transient visual loss often cannot tell the difference. A left homonymous hemianopia, that is, a defect occurring in the left visual field of each eye, may be reported by the naïve patient as loss of vision in the left eye. Patients have difficulty discriminating a homonymous hemianopia in both eyes from monocular blanking out of vision. This distinction is important because the mechanisms underlying visual loss are different in the eye and in the brain. Transient binocular visual loss is caused by migraine, a cerebral cortical process characterized by positive visual phenomena followed by headache. It is based on the physiologic mechanism called spreading depression. Spreading depression has been reproduced in vitro in the avascular retinas of frogs and chicks, but it has never been demonstrated in the perfused mammalian retina.1 Accordingly, migraine does not belong in the differential diagnosis of TMVL, and the only migraine that belongs in the differential of transient binocular visual loss is a cortical, not a retinal, process. For reasons set forth fully elsewhere, "retinal migraine" should not be included in our diagnostic vocabulary.2
|Jacqueline M.S. Winterkorn, PhD, MD, is clinical professor of ophthalmology in neurology and neuroscience, clinical professor of ophthalmology, and attending ophthalmology at Weill Cornell Medical College in New York. Dr. Winterkorn has no financial interests in any products mentioned in this article.|
Monocular visual loss has a lengthy differential diagnosis (see Table), including even blepharospasm and dry eye causing chronic blurring, and momentary visual obscurations reported with papilledema and angle closure. The description given by this patient of his visual loss lasting about 5 minutes is typical of emboli, optic disc drusen, hypercoagulability, and vasospasm. Common to all the underlying pathologies is low perfusion of the retina. Visual loss may involve the whole eye or a sector of the visual field. Although passing emboli can be experienced as bright spots seen before the visual loss, zigzag lines and other positive phenomena are absent. The similarity of these causes is that they all produce ischemia in the vascular bed of the retina. Whether embolic or vasospastic, the presentation of the patient is similar in timing. Several of the diagnostic entities produce TMVL lasting seconds, eg, papilledema and angle closure. Among the most important causes to rule out are emboli and hypercoagulability, because the underlying diagnoses may be life-threatening and are potentially treatable. Inspection of the retina occasionally discloses diagnostic embolic material. Emboli, drusen, and hypercoagulability must be ruled out before diagnosing vasospasm as the primary cause of visual loss (Figure 1).3
The source of an embolus may be determined by its appearance and ancillary testing. Calcific emboli originate in the heart and great vessels. They are grayish blobs that tend to lodge within a disc diameter of the disc itself, producing a permanent blockage in the retinal circulation and corresponding visual field. Over time, shunt vessels may form, bypassing the calcific embolus. When such emboli are found, transesophageal echocardiography should be performed looking for a new or recurrent calcific lesion.
Hollenhorst plaques are cholesterol crystals from plaque in the carotid artery. They can also be seen in the vessels close to the disc, usually at branch points, and most such emboli move from bifurcation to bifurcation and pass out of the retina. They may induce vasospasm and block the narrowed vessel leading to permanent visual field loss. Their occurrence would prompt carotid duplex scanning. But even if carotid occlusive disease is documented, endarterectomy is rarely better than medical management, including taking aspirin and lowering cholesterol.
Fibrin-platelet thrombi are only rarely visualized because they form, rapidly traverse, and pass out of the retina, and they only occasionally will be seen blocking retinal circulation. An extensive hypercoagulability workup is justified, looking for antiphospholipid antibodies, protein S, protein C, venereal disease, antinuclear antibody (ANA), factor 5 Leiden and homocysteine, complete blood count, and prothrombin time and partial thromboplastin time. Such a defect would cause thrombin-platelet blockages that might go unseen even by a careful examiner. Visual fields are important and fundus photographs may reveal what examination overlooked.
If the diagnosis is not obvious on examination, demonstrating emboli or drusen, the caliber of the vessels must be carefully assessed. In this case, the retina appeared completely normal, as were the results from visual field tests, blood tests (hypercoagulability workup, C-reactive protein, ANA, Westergren erythrocyte sedimentation rate), transesophageal echocardiogram, carotid duplex scan, magnetic resonance imaging, and intraocular pressure. Unless the observer is fortunate enough to observe the retina during an attack, primary vasospasm may remain a diagnosis of exclusion. There are, however, a half-dozen published photographs of vasospasm,4,5 and the phenomenon is well described: narrowing of arteriolar and venous blood columns, followed by retinal blanching.
Figure 1. Vasospasm in the retina of a similar patient during an attack of transient monocular visual loss, showing constriction of retinal arterioles and veins, rouleaux formation in the veins typical of low flow.
If the source of an embolus can be located, appropriate valvular surgery or anticoagulation can be considered. Many emboli probably pass through the retinal circulation without causing manifest blockage. However, when an embolus induces vasoconstriction or vasospasm, it may block the circulation. In patients with optic disc drusen, anomalous blood vessels may predispose to vasospasm and visual loss. In elderly patients, vasospasm may be induced by atherosclerosis.3 Thus, vasospasm plays a contributory role in expressing underlying embolic, thrombotic, and ischemic disease. When no underlying risk is shown, primary vasospasm may be diagnosed. In any case, vasospasm can be treated by administration of a calcium channel blocker (nifedipine [Procardia, Pfizer] or verapamil) (Figure 2).5
Retinal vasospasm as an explanation for TMVL fell out of favor in the mid-1950s after C. Miller Fisher6 observed retinal emboli passing through the retinal vascular bed and proposed them as the cause of transient visual loss. Even when vasospasm was seen, it was thought of little importance and TMVL was ascribed to occult embolization. For more than 40 years, TMVL was attributed to emboli in the retina. Only during the past 15 years has vasospasm been included in the differential diagnosis of TMVL.
Figure 2. Reperfusion of the same retina after treatment with calcium channel blocker.
Recently, vasospasm has come into favor as a primary cause of visual occlusion and even more frequently as a contributory cause in patients with blood vessels damaged by atherosclerosis, emboli, drusen, or hypertension. In such patients, the attacks of visual loss are frequent but can be controlled by calcium channel blockers such as nifedipine and verapamil.5 In patients who had been controlled on calcium channel blocker and then had discontinued medication and experienced further attacks, reinstitution of the calcium channel blocker was successful in blocking vasospasm. This patient was diagnosed with primary vasospasm and successfully treated with nifedipine. RP