Development, comparison, and internal validation of prediction models to determine the visual prognosis of patients with open globe injuries using machine learning approaches.

INTRODUCTION: Open globe injuries (OGI) represent a main preventable reason for blindness and visual impairment, particularly in developing countries. The goal of this study is evaluating key variables affecting the prognosis of open globe injuries and validating internally and comparing different machine learning models to estimate final visual acuity. MATERIALS AND METHODS: We reviewed three hundred patients with open globe injuries receiving treatment at Khatam-Al-Anbia Hospital in Iran from 2020 to 2022. Age, sex, type of trauma, initial VA grade, relative afferent pupillary defect (RAPD), zone of trauma, traumatic cataract, traumatic optic neuropathy (TON), intraocular foreign body (IOFB), retinal detachment (RD), endophthalmitis, and ocular trauma score (OTS) grade were the input features. We calculated univariate and multivariate regression models to assess the association of different features with visual acuity (VA) outcomes. We predicted visual acuity using ten supervised machine learning algorithms including multinomial logistic regression (MLR), support vector machines (SVM), K-nearest neighbors (KNN), naïve bayes (NB), decision tree (DT), random forest (RF), bagging (BG), adaptive boosting (ADA), artificial neural networks (ANN), and extreme gradient boosting (XGB). Accuracy, positive predictive value (PPV), recall, F-score, brier score (BS), Matthew correlation coefficient (MCC), receiver operating characteristic (AUC-ROC), and calibration plot were used to assess how well machine learning algorithms performed in predicting the final VA. RESULTS: The artificial neural network (ANN) model had the best accuracy to predict the final VA. The sensitivity, F1 score, PPV, accuracy, and MCC of the ANN model were 0.81, 0.85, 0.89, 0.93, and 0.81, respectively. In addition, the estimated AUC-ROC and AUR-PRC of the ANN model for OGI patients were 0.96 and 0.91, respectively. The brier score and calibration log-loss for the ANN model was 0.201 and 0.232, respectively. CONCLUSION: As classic and ensemble ML models were compared, results shows that the ANN model was the best. As a result, the framework that has been presented may be regarded as a good substitute for predicting the final VA in OGI patients. Excellent predictive accuracy was shown by the open globe injury model developed in this study, which should be helpful to provide clinical advice to patients and making clinical decisions concerning the management of open globe injuries.

Effect of oral vitamin D supplementation on dry eye disease patients with vitamin D deficiency.

CLINICAL RELEVANCE: Dry eye disease (DED) and vitamin D deficiency are both common clinical entities. Vitamin D has been reported to play a significant role in ocular surface homeostasis. BACKGROUND: This study aimed to determine the therapeutic effect of oral vitamin D supplementation in dry eye disease patients with vitamin D deficiency. METHODS: A randomized, controlled study was completed in 100 patients with dry eye disease and concurrent vitamin D deficiency (< 20 ng/ml). Participants were randomly allocated to 8 weeks of oral vitamin D supplementation with both groups receiving conventional dry eye treatment with artificial tears. Schirmer's, tear break-up time (TBUT) and osmolarity tests were measured before and after eight weeks of treatment. RESULTS: The mean age of participants was 36.8 ± 8.56 years in the treatment group (n: 50) and 34.8 ± 10.13 year in the control group (n: 50). After eight weeks of treatment the mean differences in Schirmer's, TBUT and tear osmolarity were 2.38 ± 1.55 mm, 3.95 ± 1.48 s and -16.9 ± 6.28 mOsm/L, respectively in the treatment group, and 0.7 ± 0.86 mm, 0.92 ± 1.57s and -3.34 ± 2.0 mOsm/L respectively in the control group (p < 0.001 for all parameters). The treatment group demonstrated a more significant improvement than the control group in Schirmer's, TBUT and osmolarity values (p < 0.001). CONCLUSION: Vitamin D supplementation as an adjuvant to routine dry eye treatment improves ocular surface hemostasis parameters, results in better tear stability and a more improved tear osmolarity in patients with vitamin D deficiency.

First case of fungal keratitis due to Aspergillus minisclerotigenes in Iran.

BACKGROUND AND PURPOSE: Herein, we report the first case of fungal keratitis due to Aspergillus minisclerotigenes in a 68-year-old rural woman admitted to the Ophthalmology Center of Khatam-Al-Anbia Hospital in Mashhad, northeast of Iran. CASE REPORT: The patient presented with severe pain, burning, foreign body sensation, and reduced vision in her right eye. She had long-term uncontrolled diabetes and was not able to close her eye due to an anatomical problem with the eyelid. The cornea smear sample was cultured, and the fungus was initially identified as Aspergillus flavus. The isolated strain was further identified by sequencing a part of the calmodulin gene as A. minisclerotigenes. The patient did not respond to any antifungal treatments (e.g., amphotericin B and voriconazole drops, and fluconazole 300 mg/day); therefore, she was eventually subjected to corneal transplantation surgery. CONCLUSION: Fungal keratitis can be caused by the less common species. The reliable identification of the causative agents can be accomplished by the implementation of molecular methods.

Modified deep anterior lamellar keratoplasty for the treatment of advanced keratoconus with steep corneal curvature to help in eliminating the wrinkles in the Descemet's membrane.

AIMS: To determine the clinical efficacy of modified deep anterior lamellar keratoplasty (DALK) for the treatment of advanced-stage keratoconus with steep curvature. MATERIALS AND METHODS: In this interventional, non-comparative case series, 30 patients with advanced stages of keratoconus and curvature of more than 60 D underwent a modified DALK procedure. In this technique, after big-bubble formation, posterior stromal lamella was cut and removed 5 mm centrally (baring Descemet's membrane completely) with posterior stromal layer remaining peripherally. RESULTS: The study included 30 eyes (30 patients with a mean ± SD age of 25 ± 5.4 years). The follow-up examination was performed for all participants up to 12 months after the surgery. The mean uncorrected visual acuity (UCVA) increased from 20/800 before the surgery to a subsequent 3/10 (P = 0.12). Likewise, best spectacle corrected visual acuity (BSCVA) improved, reaching 6/10 postoperatively (former quantity 20/200) (P = 0.18). In addition, mean keratometry and keratometric astigmatism managed to achieve considerable improvement, from 58.8 ± 5.4 D to 46.5 ± 2.1 D and 7.8 ± 2.1 to 4.54 ± 1.54 D, respectively (P = 0.52). Descemet's membrane wrinkling was not seen in any patient postoperatively. CONCLUSIONS: This technique is effective in restoring acceptable vision and corneal regularity in advanced cases of keratoconus with a curvature more than 60 D who are also at risk of Descemet's membrane wrinkling after DALK. Therefore, this procedure could prevent from Descemet membrane wrinkling in such cases.

Effect of topical diclofenac on postoperative photorefractive keratectomy pain: a randomized, controlled trial.

BACKGROUND: A major disadvantage of photorefractive keratectomy is postoperative pain and discomfort. This study aims to evaluate whether topical diclofenac 0.1% therapy adds any extra benefit to systemic diclofenac in controlling pain after photorefractive keratectomy. DESIGN: Prospective randomized, double-masked clinical trial applied in the Khatam-al-Anbia Eye Hospital, Mashhad University of Medical Sciences, Mashhad, Iran. PARTICIPANTS: Sixty-eight individuals (age range, 18-35 years) having bilateral photorefractive keratectomy for myopic correction with or without astigmatism. METHODS: All patients received 100 mg of extended-release oral diclofenac two times (12-h interval) before photorefractive keratectomy surgery. Postoperatively, patients were randomized into the case and control groups. Case group was administered the diclofenac 0.1% drop, one drop 2 h preoperatively and four times daily postoperatively for 3 days, whereas the control group was given artificial tears instead. MAIN OUTCOME MEASURES: Pain level, photophobia and functional activity were evaluated by the patient after operation. RESULTS: Two days after surgery, no statistically significant difference in pain level was noted between both groups. In addition, photophobia and functional activity was not different between the case and control groups. Eyelid oedema and conjunctival injection levels were significantly higher in the control group (P < 0.001). CONCLUSIONS: Administration of topical diclofenac to a standardized postoperative pain regime did not alleviate post-photorefractive keratectomy pain, although it is effective in local signs like eyelid oedema and conjunctival injection. It can be concluded that oral diclofenac is sufficient in the management of post-photorefractive keratectomy pain and addition of the topical diclofenac is unnecessary.

A simple way to clear the media for vitrectomy in eyes with corneal edema.

Herein we introduce a simple approach for clearing an edematous cornea during vitreoretinal surgery in eyes with decompensated corneal endothelium, allowing the surgeon to postpone penetrating keratoplasty. This technique was performed in 3 eyes by filling the anterior chambers with air or silicone oil, and sufficiently cleared the media for completion of vitrectomy. This simple technique enables completion of the vitrectomy without a temporary keratoprosthesis and penetrating keratoplasty in eyes with corneal edema due to endothelial decompensation.

Non-endoscopic Mechanical Endonasal Dacryocystorhinostomy.

To circumvent the disadvantages of endoscopic dacryocystorhinostomy such as small rhinostomy size, high failure rate and expensive equipment, we hereby introduce a modified technique of non-endoscopic mechanical endonasal dacryocystorhinostomy (NE-MEDCR). Surgery is performed under general anesthesia with local decongestion of the nasal mucosa. A 20-gauge vitrectomy light probe is introduced through the upper canaliculus until it touches the bony medial wall of the lacrimal sac. While directly viewing the transilluminated target area, a nasal speculum with a fiber optic light carrier is inserted. An incision is made vertically or in a curvilinear fashion on the nasal mucosa in the lacrimal sac down to the bone using a Freer periosteum elevator. Approximately 1 to 1.5 cm of nasal mucosa is removed with Blakesley forceps. Using a lacrimal punch, the thick bone of the frontal process of the maxilla is removed and the inferior half of the sac is uncovered. The lacrimal sac is tented into the surgical site with the light probe and its medial wall is incised using a 3.2 mm keratome and then excised using the Blakesley forceps. The procedure is completed by silicone intubation. The NE-MEDCR technique does not require expensive instrumentation and is feasible in any standard ophthalmic surgical setting.

Peripheral 360 degrees retinectomy in complex retinal detachment.

PURPOSE: To report the functional and anatomical results and complications of 360 degrees peripheral retinectomy for management of complicated retinal detachment. METHODS: Patients with complicated retinal detachment underwent pars plana vitrectomy, 360 degrees retinectomy, intraoperative endolaser, and internal tamponade with silicone oil. Postoperative visual acuity, intraocular pressure, retinal status, need for reoperation, and complications are presented. RESULTS: Twenty eyes of 19 patients with a mean age of 32.4 years (8-75 years) underwent pars plana vitrectomy and 360 degrees peripheral retinectomy for complicated retinal detachment due to anterior proliferative vitreoretinopathy, unstable edge of retinal break, anterior hyaloidal fibrovascular proliferation, retinal incarceration in scleral wound, and 300 degrees giant retinal tear. Intraoperative reattachment was achieved in 18 eyes. Mean postoperative follow-up time was 24.2 months (2-70 months). Retina was attached in 14 eyes (70%) in the last visit. Eight eyes (40%) had 5/200 or greater visual acuity. Preoperative and postoperative visual acuities did not have significant correlation (Spearman correlation coefficient = 0.291). There was no relation between diagnosis and anatomical outcome (P > 0.2). CONCLUSION: Relaxing peripheral 360 degrees retinectomy is an effective procedure for flattening the retina in complicated retinal detachments when no other option is available.

Non-endoscopic endonasal dacryocystorhinostomy--technique, indications, and results.

PURPOSE: To describe the technique of non-endoscopic endonasal dacryocystorhinostomy (NEN-DCR), and its indications and results. MATERIALS AND METHODS: In a prospective, non-randomized interventional case series, all consecutive cases presenting with epiphora between 2004 and 2006 were enrolled. Adult patients were divided into three subgroups: chronic nasolacrimal duct obstruction (NLDO), NLDO with dacryocystitis (NLDO-DC), and recurrent NLDO with previous failed external DCR (REV-DCR). All procedures were performed by one surgeon (first author). An endonasal DCR was performed through a nasal speculum without the use of an endoscope. Success was measured by both improvement of the epiphora and patency of the lacrimal system during irrigation. RESULTS: Ninety-five patients (24 men, 71 women) underwent 99 NENDCR procedures; 54% of cases had NLDO, 32% had NLDO-DC, and 14% were REV-DCR. Mean duration of surgery was 30 minutes, and the average amount of intraoperative bleeding was 12 ml. After a minimum follow-up of 6 months, success was achieved in 96% of all patients, 94% in the NLDO group, 97% in the NLDO-DC group, and 92% in the REV-DCR group. The failure rate was 4% overall. Two patients failed from the NLDO group, one patient from the NLDO-DC group, and one patient from the REV-DCR group. No significant late complications were detected. CONCLUSION: The success rate of NEN-DCR compares favorably with external DCR. The technique is also useful in cases of NLDO-DC and DCR-REV.

Isolated orbital endodermal sinus tumor.

PURPOSE: To describe the clinical, serologic, histopathologic, and immunohistochemical findings of an isolated endodermal sinus tumor of the orbit in a 1-year-old boy. METHODS: A retrospective case report and literature review. RESULTS: The patient was managed by the ophthalmology and oncology services. After the diagnosis was established by serologic analysis and histopathology, the patient underwent chemotherapy. CONCLUSIONS: Endodermal sinus tumor is a malignant germ-cell tumor that usually involves the gonadal tissue. Primary extragonadal sites occur rarely, and include the orbit. Treatment options include surgical removal and chemotherapy.