Late recurrence of retinal detachment

Deep Learning for prediction of late recurrence of retinal detachment using preoperative and postoperative ultra-wide field imaging

PURPOSE

To elaborate a deep learning (DL) model for automatic prediction of late recurrence (LR) of rhegmatogenous retinal detachment (RRD) using pseudocolor and fundus autofluorescence (AF) ultra-wide field (UWF) images obtained preoperatively and postoperatively.

MATERIALS AND METHODS

We retrospectively included patients >18 years who underwent either scleral buckling (SB) or pars plana vitrectomy (PPV) for primary or recurrent RRD with a post-operative follow-up >2 years. Records of RRD recurrence between 6 weeks and 2 years after surgery served as a ground truth for the training of the deep learning (DL) models. Four separate DL models were trained to predict LR within the 2 postoperative years (binary outputs) using, respectively, UWF preoperative and postoperative pseudocolor images and UWF preoperative and postoperative AF images.

RESULTS

A total of 412 eyes were included in the study (332 eyes treated with PPV and 80 eyes with SB). The mean follow-up was 4.0 ± 2.1 years. The DL models based on preoperative and postoperative pseudocolor UWF imaging predicted recurrence with 85.6% (sensitivity 86.7%, specificity 85.4%) and 90.2% accuracy (sensitivity 87.0%, specificity 90.8%) in PPV-treated eyes, and 87.0% (sensitivity 86.7%, specificity 87.0%) and 91.1% (sensitivity 88.2%, specificity 91.9%) in SB-treated eyes, respectively. The DL models using preoperative and postoperative AF-UWF imaging predicted recurrence with 87.6% (sensitivity 84.0% and specificity 88.3%) and 91.0% (sensitivity 88.9%, specificity 91.5%) accuracy in PPV eyes, and 86.5% (sensitivity 87.5%; specificity 86.2%) and 90.6% (sensitivity 90.0%, specificity 90.7%) in SB eyes, respectively. Among the risk factors detected with visualisation methods, potential novel ones were extensive laser retinopexy and asymmetric staphyloma.

CONCLUSIONS

DL can accurately predict the LR of RRD based on UWF images (especially postoperative ones), which can help refine follow-up strategies. Saliency maps might provide further insight into the dynamics of RRD recurrence.

LATE RECURRENCE OF RETINAL DETACHMENT: Incidence, Clinical Features, and Surgical Outcomes

PURPOSE

To describe and evaluate demographics, clinical features, prognostic factors, rate of success of surgery, incidence, and visual outcomes in patients with a late recurrence of rhegmatogenous retinal detachment over a 10-year period at a large tertiary referral eye center.

METHODS

A retrospective, observational case series of patients with late recurrence of retinal detachment, defined as redetachment after at least six months of total reattachment in non-proliferative vitreoretinopathy (PVR) rhegmatogenous retinal detachment, after pars plana vitrectomy (PPV) surgery with gas tamponade.

RESULTS

Thirty-nine patients had a late recurrence of rhegmatogenous retinal detachment of 16,396 rhegmatogenous retinal detachment operations. The mean of time between the first retinal detachment (RD) surgery and redetachment was 122.7 (SD 115) weeks. On presentation with late recurrence, 72% of eyes were pseudophakic and 64% were macula-off. In 28 eyes, small breaks were found. Thirty-eight percent had established PVR (PVR-C in 80%). Ninety-five percent underwent PPV. Gas was used in 61%. The initial secondary success rate was 64%. Initial best-corrected visual acuity was 1.32 logarithm of the minimum angle of resolution (logMAR) (6/120) and final was 0.8 logMAR (6/38; P value 0.002).

CONCLUSION

Late recurrence of retinal detachment is rare. It is characterized by small retinal breaks that may be difficult to visualize. Although cases can be treated with favorable anatomical results, visual outcomes are often less good and the success rate is lower.

Risk of Tertiary, Quaternary, and Quinary Proliferative Vitreoretinopathy: Analysis of a Nationwide Database (2010-2017)

OBJECTIVE

Primary proliferative vitreoretinopathy (PVR) is established as an important cause of the failed repair of a fresh retinal detachment (RD) and the consequent need for secondary repair. However, the burden of multiple repairs beyond the initial failure has not been studied in detail. We aimed to determine the association between primary PVR and the occurrence of tertiary, quaternary, and quinary RD repairs, using a nationwide database.

DESIGN

Retrospective cohort study of insurance claims.

SUBJECTS

Cases of rhegmatogenous RD that underwent primary surgical repair.

METHODS

Cases of primary RD repair from 2010 to 2017 were categorized based on the absence (P0 group) or presence (P1 group) of primary PVR. In each group, we analyzed the frequency of subsequent RD repair procedures with concurrent PVR.

MAIN OUTCOME MEASURE

The risk of secondary and higher multiples of PVR-associated RD repair.

RESULTS

A total of 27 137 cases were included, with 24 500 (90.3%) in the P0 group and 2637 (9.7%) in the P1 group. The frequency (%) of cases ultimately requiring secondary, tertiary, quaternary, and quinary repair in P0 versus P1 was 1.88 versus 10.24 (P < 0.001), 0.26 versus 2.50 (P < 0.001), 0.07 versus 0.64 (P < 0.001), and 0.03 versus 0.08 (P = 0.272), respectively. The risk of undergoing secondary repair was higher in the P1 than in the P0 group (hazard ratio [HR], 6.02; 95% confidence interval [CI], 5.24-6.92; P < 0.001). The risk of undergoing tertiary repair was also higher in the P1 than in the P0 group (HR, 1.67; CI, 1.23-2.28; P = 0.001). There was no difference in the risk of undergoing quaternary repair between the groups (HR, 0.76; CI, 0.41-1.40; P = 0.37). Senary repairs were not detected in this dataset.

CONCLUSIONS

Primary PVR may increase the risk of requiring multiple sequential retinal reattachment surgeries beyond the initial repair failure. Retinal detachment cases with primary PVR at the initial presentation of RD were more likely to undergo secondary and tertiary repairs than cases without primary PVR. Health care claims analysis may be a useful tool to study population-based estimates for multiple recurrences of RD in cases with PVR.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found after the references.