Using SD-OCT to Detect Complete PVD
By Jean Shaw
Selected By: Andrew P. Schachat, MD
Journal Highlights
Ophthalmology Retina, February 2020
Download PDF
Hwang et al. set out to assess whether preoperative spectral-domain optical coherence tomography (SD-OCT) of the macula could accurately detect posterior vitreous detachment (PVD). They found that an accurate determination of attached vitreous can be made if the premacular bursa or posterior vitreous cortex are visualized.
For this retrospective chart review, the researchers evaluated 175 patients (175 eyes) who underwent vitrectomy surgery between Jan. 1, 2009, and Dec. 31, 2017. Two masked ophthalmologists independently graded the patients’ preoperative SD-OCT scans, and those results were compared against the treating surgeons’ intraoperative notes regarding the likelihood of PVD. Attached vitreous was identified on OCT by visualization of the posterior vitreous cortex or the premacular bursa. Complete PVD was identified by the absence of these findings and was considered a positive outcome for the purpose of the analysis.
During the grading process, 38 eyes were confirmed as having complete PVD, and 137 were assessed as attached vitreous. However, at the time of surgery, 20 of the 38 eyes with complete PVD were described as having preexisting PVD (true positives), and 18 were documented as having attached vitreous (false positives). With regard to the 137 eyes graded as attached vitreous on OCT, 129 had attached vitreous at the time of surgery (true negatives), while eight had preexisting PVD at that time (false negatives).
The sensitivity of SD-OCT for detecting complete PVD was 71%, the specificity was 88%, the positive predictive value was 53%, and the negative predictive value was 94%. The most common diagnosis was macular hole (53%); in these eyes, the sensitivity of pre-op OCT in correctly diagnosing PVD was 67%, specificity was 88%, positive predictive value was 38%, and negative predictive value was 96%.
The authors noted that the OCT scans were not obtained with the enhanced vitreous imaging technique, which aids in visualizing vitreous structures. In particular, they said, if neither the premacular bursa nor the posterior vitreous cortex is visualized, SD-OCT has a poor predictive value, and ultrasound is needed to accurately identify complete PVD. (Also see related commentary by Justis P. Ehlers, MD, in the same issue.)
The original article can be found here.