OBJECTIVE: To determine the the effectiveness of random particle motion, presented on a computer monitor, as a noninvasive test for detecting cytomegalovirus retinitis. DESIGN: A prospective masked study in which patients were asked to trace out any disturbances on a transparency placed over a computer monitor that displayed continuous random particle motion, while the patient fixated on a central spot (entoptic perimetry). SETTING: The Acquired Immunodeficiency Syndrome Ocular Research Unit at the University of California, San Diego, in La Jolla. PATIENTS: Twenty-two men with cytomegalovirus retinitis who were positive for human immunodeficiency virus, 11 men without cytomegalovirus retinitis who were positive for human immunodeficiency virus, and eight men who were negative for human immunodeficiency virus. INTERVENTION: None. MEASUREMENTS: Sensitivities and specificities were used to compare the results of entoptic perimetry with fundus photographs. RESULTS: Entoptic perimetry demonstrated a 95% sensitivity and a 95% specificity in detection of cytomegalovirus retinitis. CONCLUSION: Entoptic perimetry may be an effective and inexpensive screening test for cytomegalovirus retinitis in hospitals and community clinics.
OBJECTIVE: To determine the the effectiveness of random particle motion, presented on a computer monitor, as a noninvasive test for detecting cytomegalovirus retinitis. DESIGN: A prospective masked study in which patients were asked to trace out any disturbances on a transparency placed over a computer monitor that displayed continuous random particle motion, while the patient fixated on a central spot (entoptic perimetry). SETTING: The Acquired Immunodeficiency Syndrome Ocular Research Unit at the University of California, San Diego, in La Jolla. PATIENTS: Twenty-two men with cytomegalovirus retinitis who were positive for human immunodeficiency virus, 11 men without cytomegalovirus retinitis who were positive for human immunodeficiency virus, and eight men who were negative for human immunodeficiency virus. INTERVENTION: None. MEASUREMENTS: Sensitivities and specificities were used to compare the results of entoptic perimetry with fundus photographs. RESULTS: Entoptic perimetry demonstrated a 95% sensitivity and a 95% specificity in detection of cytomegalovirus retinitis. CONCLUSION: Entoptic perimetry may be an effective and inexpensive screening test for cytomegalovirus retinitis in hospitals and community clinics.