We would like to extend an invitation to readers to submit surgical pearls for publication in Retina Today. Please send submissions for consideration to either Dean Eliott, MD, (deliott@doheny.org), or Ingrid U. Scott, MD, MPH, (iscott@psu.edu). We're looking forward to hearing from you.
Since its approval by the US Food and Drug Administration in January 2002, more than 6,000 Stratus OCT (Carl Zeiss Meditec, Dublin, CA) systems have been installed worldwide.1 It is estimated that 37,000 daily scans are performed in the United States.1 Based on these numbers, it is clear that OCT has become a valuable adjunct to clinical practice. For example, OCT has been shown to have a higher sensitivity than stereoscopic biomicroscopy for the detection of diabetic macular edema2 and provide a more reliable diagnosis than can be achieved with clinical examination in the early stages of macular holes.3 Additionally, OCT has introduced new clinical concepts—such as stage "0" macular hole.4
Despite its enormous utility and ease of use, however, the Stratus OCT can deliver inaccurate results if the clinician is unaware of the basic software functions and algorithms. Common errors in the daily use of the Stratus can be classified into two major groups: acquisition errors and interpretation errors.
ACQUISITION ERRORS
Poor scan acquisition can be responsible for an artifact in approximately 11% of topographic maps.5 One of the major pitfalls of the Stratus OCT is that image acquisition is dependent on the patient's fixation. This can affect the precision of measurements and can compromise comparison of baseline OCT images with OCT images at subsequent follow-up scans (Figure 1). If the fellow eye maintains fixation, the use of the external fixation LED can be helpful. In cases where the normal anatomy is altered, the expertise of the examiner plays a major role in the correct positioning of the line scan. The software automatically chooses the retinal boundaries to calculate retinal thickness. If the anatomy is altered, the boundaries are often misplaced, leading to errors in measurement. In these cases, the operator should enter the analysis screen and remove the white line (by unchecking the Layer-On box). Calipers should be switched on (by checking the appropriate box) and placed on the appropriate areas to make a manual measurement.
During image acquisition, a high-quality scan must show a signal strength of five add units or more. Media opacities can affect the strength of the signal, resulting in an inaccurate exam. Good signal strength does not ensure a reliable OCT analysis, and the clinician should look for any disparity in the report, as the Stratus software version 4.0 can identify the presence of artifacts (Figure 2).
INTERPRETATION ERRORS
The analyzing software for retinal scans includes an automated correction for movement and an algorithm for retinal thickness measurements, both of which can produce errors during the analysis of the scans.
The purpose of the movement correction is to straighten the retinal pigment epithelium (RPE) layer in case of fluctuations related to eye movement during the scan. This software can mask certain alterations of the RPE related to different pathologies. In age-related macular degeneration, the movement correction can hide the presence of drusen and alter the retinal contour (Figure 3), or it can flatten retinal pigment epithelial detachments (PED; Figure 4). In such cases, the usefulness of the OCT relies on the detailed information that can be obtained through the OCT images. To override the morphological alterations in the RPE of the movement correction software, we recommend evaluation of the unprocessed images found under the Scan Profile analysis or the Scan Selection section (Figure 5).
In central serous chorioretinopathy the movement correction can also flatten the the typical rounded PED or even hide PEDs if they are small (Figure 6).
Measurements of the retinal thickness algorithm are based on the automated capability of the OCT to identify the internal limiting membrane and the hyperreflective band of the RPE-choriocapillaris. Errors related to this algorithm are remarkably common but are generally detected by the software.6,7 Conditions that are characterized by the presence of subretinal fluid are related to more severe errors than those with retinal cysts or retinal vascular diseases.7 Sadda and colleagues8 reported the use of customized software in which the manual measurements were consistent with those delivered with the automated Stratus analysis.
The presence of vitreoretinal traction may cause a failure in the localization of the internal limiting membrane with misplacement of the inner boundary on the zone of vitreous traction (Figure 7). The presence of a dense retinal hemorrhage can obscure the RPE-choriocapillaris band producing an incorrect interpretation of the outer boundary measurement (Figure 8). Images with poor signal strength are also prone to misdirection of the automated boundary layers of the algorithm.
CONCLUSION
Although errors are common in the image analysis of the Stratus OCT, the current software version 4.0 can detect and warn of their presence. The examiner must search for any incongruent information. In many pathologies, the information of the OCT images is very useful; however, images of entities with alterations of the RPE morphology are best evaluated without any analysis protocol. Spectral-domain OCT technology has overcome many of these weaknesses, such as fixation dependence, small scanned area, and thickness measurements, but Stratus OCT can still be a valuable tool in the daily clinical practice if the examiner is aware of its normal functioning and pitfalls.
Lihteh Wu, MD, is Consulting Surgeon in the Department of Ophthalmology, Vitreo-Retinal Section, Instituto De Cirugia Ocular, Costa Rica.
Teodoro Evans, MD, is a retina fellow, Vitreo-Retinal Section, Instituto De Cirugia Ocular. Dr. Wu and Dr. Evans report no financial interests. Dr. Wu can be reached by phone: +1 506 2256 2134; fax: +1 506 2256 2128; or lw65@cornell.edu.
Dean Eliott, MD, is Professor of Ophthalmology and Director of Clinical Affairs, Doheny Eye Institute, Keck School of Medicine at USC and is a member of the Retina Today Editorial Board. He may be reached by phone: +1 323 442 6582; fax: +1 323 442 6766; or deliott@doheny.org.
Ingrid U. Scott, MD, MPH, is Professor of Ophthalmology and Public Health Sciences, Penn State College of Medicine, Department of Ophthalmology and is a member of the Retina Today Editorial Board.. She may be reached by phone: +1 717 531 4662; or fax: +1 717 531 5475; or iscott@psu.edu.
_1773249222.png?auto=compress,format&w=75)
