When a patient is seeing stars, when is it a true ocular emergency?
Recently on the EM:RAP podcast, we discussed the common presenting complaint of visual floaters and flashes. Our discussion was prompted by a recent piece in the JAMA Rational Clinical Examination series.(1) This is a great series of articles that examines the diagnostic utility of various clinical findings when considering important diagnoses in the setting of common presenting complaints.
Floaters are a sensation of moving spots, usually gray or dark in color, across the visual field of one eye. Flashes are brief, repeated sensations of bright light, usually at the periphery of the visual field. Both are painless.
Although almost all causes of floaters and flashes are ocular, one should start the assessment by ensuring that symptoms referable to other systems are not present. Non-ocular causes of floaters and flashes include postural hypotension and migraine, both of which should be easily distinguished by the presence of other clinical features such as lightheadedness and headache, and by the fact that they typically result in bilateral eye symptoms. Unlike floaters and flashes of ocular origin, the flashing lights (scintillations) of migraine are most often colorful. Curtains or cobwebs encroaching on any part of the visual field of one eye represent retinal detachment until proven otherwise and represent a true ocular emergency. Transient ischemic attack (TIA) involving the ophthalmic artery, amaurosis fugax, also results in unilateral symptoms, typically a curtain-like visual field cut lasting seconds to minutes in the upper or lower half of the visual field of one eye.
The vast majority of floaters and flashes are caused by posterior vitreous detachment (PVD). As we age, the vitreous humor, which is essentially a bag of jelly that occupies the posterior chamber of the eye, shrinks. As the vitreous shrinks, it separates from the retina. The bending of light that occurs in these pockets of separation and the associated traction on the peripheral retina results in the perception of floaters and flashes, respectively.
PVD is extremely common. In fact, most people will experience the symptoms of PVD at some point in their lives. The key point to remember is that with time, PVD becomes more stable, so that patients with longstanding symptoms of floaters and flashes (months to years) are of much less concern. In the acute phase (days to weeks), however, the tractional forces on the areas of retina that are newly unsupported by underlying vitreous may be sufficient to lead to a retinal tear. Retinal tears may in turn lead to retinal detachment and permanent visual loss if not promptly identified and treated by an ophthalmologist. Upon receiving these patients, ophthalmologists will generally perform a dilated examination and indirect ophthalmoscopy, which is able to detect tears at the periphery of the retina that may be otherwise undetectable.
In the study by Hollands et al., data from 17 different studies of patients with acute onset floaters and flashes of suspected ocular origin were analyzed. In this group of over one thousand patients, the overall incidence of retinal tear was 14%. Of all of the findings on history and examination, vitreous hemorrhage or pigment (so-called “tobacco dust”) on direct ophthalmoscopy and/or slit lamp examination was most strongly predictive of retinal tear. However, most EPs do not have the experience necessary to reliably detect these findings. More important to the EP was the strong predictive value of a subjective description of decreased visual acuity. Patients who reported a decrease in their vision in addition to floaters and flashes were several times more likely to have a retinal tear.
The most important message from the study was that none of clinical findings studied were able to adequately to reduce the probability of retinal tear to levels that most of us would consider acceptable. In other words, ophthalmological consultation is unavoidable – the only question that remains is how urgently does this need to occur? The authors suggest that in patients with high-risk findings (e.g. decreased visual acuity, either subjective or objective, or any finding on slit-lamp examination or direct ophthalmoscopy) consultation should occur on the same day. Those with frank signs of detachment, such as a curtain of darkness encroaching on the visual field or the appearance of a billowing retina on ophthalmoscopy, require emergent consultation, as minutes may matter in these cases. For patients without any other findings, they suggest that follow-up should occur within 1-2 weeks, with the caveat that if there is any progression or change in symptoms that they should return immediately. Contacting an ophthalmologist to determine the urgency of the consultation is probably the most prudent course of action, however, especially if the EP feels uncomfortable with his or her direct ophthalmoscopy or slit lamp skills.
One factor that the authors of the paper, who are ophthalmologists, did not consider was the emergence of bedside ultrasound as used by emergency and primary care physicians. Detachments that may be clinically difficult to detect are sometimes easily seen on ocular ultrasound. This would be one finding that would make an ophthalmology consult much more emergent.
One last thing: patients with longstanding diabetes are different. In these patients, vitreous hemorrhage from friable retinal vessels may mimic PVD. Urgent consultation is appropriate in these patients as well, and again, its timing is best discussed with an ophthalmologist.
1. Hollands H, et al. Acute-Onset Floaters and Flashes: Is this Patient at Risk for Retinal Detachment? JAMA 2009;302(20):2243-2249.
Dr. Swadron is currently Vice-Chair for Education in the Department of Emergency Medicine at the Los Angeles County/USC Medical Center in Los Angeles. He is an Associate Professor of Clinical Emergency Medicine at the Keck School of Medicine of the University of Southern California. EM:RAP (Emergency Medicine: Reviews and Perspectives) is a monthly audio program that can be found at www.EMRAP.org