Optimising vitrectomy operation note coding with machine learning.

BACKGROUND: The accurate encoding of operation notes is essential for activity-based funding and workforce planning. The aim of this project was to evaluate the procedural coding accuracy of vitrectomy and to develop machine learning, natural language processing (NLP) models that may assist with this task. METHODS: This retrospective cohort study involved vitrectomy operation notes between a 21-month period at the Royal Adelaide Hospital. Coding of procedures were based on the Medicare Benefits Schedule (MBS)-the Australian equivalent to the Current Procedural Terminology (CPT®) codes used in the United States. Manual encoding was conducted for all procedures and reviewed by two vitreoretinal consultants. XGBoost, random forest and logistic regression models were developed for classification experiments. A cost-based analysis was subsequently conducted. RESULTS: There were a total of 1724 procedures with individual codes performed within 617 vitrectomy operation notes totalling $1 528 086.60 after manual review. A total of 1147 (66.5%) codes were missed in the original coding that amounted to $736 539.20 (48.2%). Our XGBoost model had the highest classification accuracy (94.6%) in the multi-label classification for the five most common procedures. The XGBoost model was the most successful model in identifying operation notes with two or more missing codes with an AUC of 0.87 (95% CI 0.80-0.92). CONCLUSIONS: Machine learning has been successful in the classification of vitrectomy operation note encoding. We recommend a combined human and machine learning approach to clinical coding as automation may facilitate more accurate reimbursement and enable surgeons to prioritise higher quality clinical care.

Safety Profile of Slit-Lamp-Delivered Retinal Laser Photobiomodulation.

Purpose: Photobiomodulation (PBM) refers to therapeutic irradiation of tissue with low-energy, 630- to 1000-nm wavelength light. An increasing body of evidence supports a beneficial effect of PBM in retinal disorders. To date, most studies have utilized light-emitting diode irradiation sources. Slit-lamp-mounted retinal lasers produce a coherent beam that can be delivered with precisely defined dosages and predetermined target area; however, the use of retinal lasers raises safety concerns that warrant investigation prior to clinical application. In this study, we determined safe dosages of laser-delivered PBM to the retina. Methods: A custom-designed, slit-lamp-delivered, 670-nm, red/near-infrared laser was used to administer a range of irradiances to healthy pigmented and non-pigmented rat retinas. The effects of PBM on various functional and structural parameters of the retina were evaluated utilizing a combination of electroretinography, Spectral Domain Optical Coherence (SD-OCT), fluorescein angiography, histology and immunohistochemistry. Results: In non-pigmented rats, no adverse events were identified at any irradiances up to 500 mW/cm2. In pigmented rats, no adverse events were identified at irradiances of 25 or 100 mW/cm2; however, approximately one-third of rats that received 500 mW/cm2 displayed very localized photoreceptor damage in the peripapillary region, typically adjacent to the optic nerve head. Conclusions: A safety threshold exists for laser-delivered PBM in pigmented retinas and was identified as 500 mW/cm2 irradiance; therefore, caution should be exercised in the dosage of laser-delivered PBM administered to pigmented retinas. Translational Relevance: This study provides important data necessary for clinical translation of laser-delivered PBM for retinal diseases.

Preclinical and clinical studies of photobiomodulation therapy for macular oedema.

AIMS/HYPOTHESIS: Diabetic macular oedema (DME) is the leading cause of visual impairment in people with diabetes. Intravitreal injections of vascular endothelial growth factor inhibitors or corticosteroids prevent loss of vision by reducing DME, but the injections must be given frequently and usually for years. Here we report laboratory and clinical studies on the safety and efficacy of 670 nm photobiomodulation (PBM) for treatment of centre-involving DME. METHODS: The therapeutic effect of PBM delivered via a light-emitting diode (LED) device was tested in transgenic mice in which induced Müller cell disruption led to photoreceptor degeneration and retinal vascular leakage. We also developed a purpose-built 670 nm retinal laser for PBM to treat DME in humans. The effect of laser-delivered PBM on improving mitochondrial function and protecting against oxidative stress was studied in cultured rat Müller cells and its safety was studied in pigmented and non-pigmented rat eyes. We then used the retinal laser to perform PBM in an open-label, dose-escalation Phase IIa clinical trial involving 21 patients with centre-involving DME. Patients received 12 sessions of PBM over 5 weeks for 90 s per treatment at a setting of 25, 100 or 200 mW/cm2 for the three sequential cohorts of 6-8 patients each. Patients were recruited from the Sydney Eye Hospital, over the age of 18 and had centre-involving DME with central macular thickness (CMT) of >300 µm with visual acuity of 75-35 Log minimum angle of resolution (logMAR) letters (Snellen visual acuity equivalent of 20/30-20/200). The objective of this trial was to assess the safety and efficacy of laser-delivered PBM at 2 and 6 months. The primary efficacy outcome was change in CMT at 2 and 6 months. RESULTS: LED-delivered PBM enhanced photoreceptor mitochondrial membrane potential, protected Müller cells and photoreceptors from damage and reduced retinal vascular leakage resulting from induced Müller cell disruption in transgenic mice. PBM delivered via the retinal laser enhanced mitochondrial function and protected against oxidative stress in cultured Müller cells. Laser-delivered PBM did not damage the retina in pigmented rat eyes at 100 mW/cm2. The completed clinical trial found a significant reduction in CMT at 2 months by 59 ± 46 µm (p = 0.03 at 200 mW/cm2) and significant reduction at all three settings at 6 months (25 mW/cm2: 53 ± 24 µm, p = 0.04; 100 mW/cm2: 129 ± 51 µm, p < 0.01; 200 mW/cm2: 114 ± 60 µm, p < 0.01). Laser-delivered PBM was well tolerated in humans at settings up to 200 mW/cm2 with no significant side effects. CONCLUSIONS/INTERPRETATION: PBM results in anatomical improvement of DME over 6 months and may represent a safe and non-invasive treatment. Further testing is warranted in randomised clinical trials. TRIAL REGISTRATION: ClinicalTrials.gov NCT02181400 Graphical abstract.

Improvement in inner retinal function in glaucoma with nicotinamide (vitamin B3) supplementation: A crossover randomized clinical trial.

IMPORTANCE: Retinal ganglion cells endure significant metabolic stress in glaucoma but maintain capacity to recover function. Nicotinamide, a precursor of NAD+ , is low in serum of glaucoma patients and its supplementation provides robust protection of retinal ganglion cells in preclinical models. However, the potential of nicotinamide in human glaucoma is unknown. BACKGROUND: To examine the effects of nicotinamide on inner retinal function in glaucoma, in participants receiving concurrent glaucoma therapy. DESIGN: Crossover, double-masked, randomized clinical trial. Participants recruited from two tertiary care centres. PARTICIPANTS: Fifty-seven participants, diagnosed and treated for glaucoma. METHODS: Participants received oral placebo or nicotinamide and reviewed six-weekly. Participants commenced 6 weeks of 1.5 g/day then 6 weeks of 3.0 g/day followed by crossover without washout. Visual function measured using electroretinography and perimetry. MAIN OUTCOME MEASURES: Change in inner retinal function, determined by photopic negative response (PhNR) parameters: saturated PhNR amplitude (Vmax), ratio of PhNR/b-wave amplitude (Vmax ratio). RESULTS: PhNR Vmax improved beyond 95% coefficient of repeatability in 23% of participants following nicotinamide vs 9% on placebo. Overall, Vmax improved by 14.8% [95% CI: 2.8%, 26.9%], (P = .02) on nicotinamide and 5.2% [-4.2%, 14.6%], (P = .27) on placebo. Vmax ratio improved by 12.6% [5.0%, 20.2%], (P = .002) following nicotinamide, 3.6% [-3.4%, 10.5%], (P = .30) on placebo. A trend for improved visual field mean deviation was observed with 27% improving ≥1 dB on nicotinamide and fewer deteriorating (4%) compared to placebo (P = .02). CONCLUSIONS: Nicotinamide supplementation can improve inner retinal function in glaucoma. Further studies underway to elucidate the effects of long-term nicotinamide supplementation.

Investigations Into Bioenergetic Neuroprotection of Cone Photoreceptors: Relevance to Retinitis Pigmentosa.

Recent studies suggest cone degeneration in retinitis pigmentosa (RP) may result from intracellular energy depletion. We tested the hypothesis that cones die when depleted of energy by examining the effect of two bioenergetic, nutraceutical agents on cone survival. The study had three specific aims: firstly, we, studied the neuroprotective efficacies of glucose and creatine in an in vitro model of RP. Next, we utilized a well-characterized mouse model of RP to examine whether surviving cones, devoid of their inner segments, continue to express genes vital for glucose, and creatine utilization. Finally, we analyzed the neuroprotective properties of glucose and creatine on cone photoreceptors in a mouse model of RP. Two different bioenergy-based therapies were tested in rd1 mice: repeated local delivery of glucose and systemic creatine. Optomotor responses were tested and cone density was quantified on retinal wholemounts. The results showed that glucose supplementation increased survival of cones in culture subjected to mitochondrial stress or oxidative insult. Despite losing their inner segments, surviving cones in the rd1 retina continued to express the various glycolytic enzymes. Following a single subconjunctival injection, the mean vitreous glucose concentration was significantly elevated at 1 and 8 h, but not at 16 h after injection; however, daily subconjunctival injection of glucose neither enhanced spatial visual performance nor slowed cone cell degeneration in rd1 mice relative to isotonic saline. Creatine dose-dependently increased survival of cones in culture subjected to mitochondrial dysfunction, but not to oxidative stress. Despite the loss of their mitochondrial-rich inner segments, cone somas and axonal terminals in the rd1 retina were strongly positive for both the mitochondrial and cytosolic forms of creatine kinase at each time point examined. Creatine-fed rd1 mice displayed enhanced optomotor responses compared to mice fed normal chow. Moreover, cone density was significantly greater in creatine-treated mice compared to controls. The overall results of this study provide tentative support for the hypothesis that creatine supplementation may delay secondary degeneration of cones in individuals with RP.

Creatine is Neuroprotective to Retinal Neurons In Vitro But Not In Vivo.

Purpose: To investigate the neuroprotective properties of creatine in the retina using in vitro and in vivo models of injury. Methods: Two different rat retinal culture systems (one containing retinal ganglion cells [RGC] and one not) were subjected to either metabolic stress, via treatments with the mitochondrial complex IV inhibitor sodium azide, or excitotoxic stress, via treatment with N-methyl-D-aspartate for 24 hours, in the presence or absence of creatine (0.5, 1.0, and 5.0 mM). Neuronal survival was assessed by immunolabeling for cell-specific antigens. Putative mechanisms of creatine action were investigated in vitro. Expression of creatine kinase (CK) isoenzymes in the rat retina was examined using Western blotting and immunohistochemistry. The effect of oral creatine supplementation (2%, wt/wt) on retinal and blood creatine levels was determined as well as RGC survival in rats treated with N-methyl-D-aspartate (NMDA; 10 nmol) or high IOP-induced ischemia reperfusion. Results: Creatine significantly prevented neuronal death induced by sodium azide and NMDA in both culture systems. Creatine administration did not alter cellular adenosine triphosphate (ATP). Inhibition of CK blocked the protective effect of creatine. Retinal neurons, including RGCs, expressed predominantly mitochondrial CK isoforms, while glial cells expressed exclusively cytoplasmic CKs. In vivo, NMDA and ischemia reperfusion caused substantial loss of RGCs. Creatine supplementation led to elevated blood and retinal levels of this compound but did not significantly augment RGC survival in either model. Conclusions: Creatine increased neuronal survival in retinal cultures; however, no significant protection of RGCs was evident in vivo, despite elevated levels of this compound being present in the retina after oral supplementation.

Retinal pigment epithelium in the pathogenesis of age-related macular degeneration and photobiomodulation as a potential therapy?

The retinal pigment epithelium (RPE) comprises a monolayer of cells located between the neuroretina and the choriocapillaries. The RPE serves several important functions in the eye: formation of the blood-retinal barrier, protection of the retina from oxidative stress, nutrient delivery and waste disposal, ionic homeostasis, phagocytosis of photoreceptor outer segments, synthesis and release of growth factors, reisomerization of all-trans-retinal during the visual cycle, and establishment of ocular immune privilege. Age-related macular degeneration (AMD) is the leading cause of blindness in developed countries. Dysfunction of the RPE has been associated with the pathogenesis of AMD in relation to increased oxidative stress, mitochondrial destabilization and complement dysregulation. Photobiomodulation or near infrared light therapy which refers to non-invasive irradiation of tissue with light in the far-red to near-infrared light spectrum (630-1000 nm), is an intervention that specifically targets key mechanisms of RPE dysfunction that are implicated in AMD pathogenesis. The current evidence for the efficacy of photobiomodulation in AMD is poor but its safety profile and proposed mechanisms of action motivate further research as a novel therapy for AMD.

Risk of Intraocular Bleeding With Novel Oral Anticoagulants Compared With Warfarin: A Systematic Review and Meta-analysis.

Importance: It is unclear if the risk of intraocular bleeding with novel oral anticoagulants differs compared with warfarin. Objective: To characterize the risk of intraocular bleeding with novel oral anticoagulants compared with warfarin. Data Sources: A systematic review and meta-analysis was undertaken in an academic medical setting. MEDLINE and ClinicalTrials.gov were searched for randomized clinical trials published up until August 2016. This search was supplemented by manual bibliography searches of identified trials and other review articles. Study Selection: Studies were eligible for inclusion if they were phase 3 randomized clinical trials, enrolled patients with atrial fibrillation or venous thromboembolism, compared a novel oral anticoagulant (dabigatran, rivaroxaban, apixaban, or edoxaban) with warfarin, and recorded event data on intraocular bleeding. Data on intraocular bleeding were pooled using inverse-variance, weighted, fixed-effects meta-analysis. Data Extraction and Synthesis: The PRISMA guidelines were used for abstracting data and assessing quality. Independent extraction was performed by 2 investigators. Main Outcomes and Measures: Intraocular bleeding events and associated risk ratio for novel oral anticoagulants compared with warfarin. Results: Twelve trials investigating 102 627 patients were included. Randomization to novel oral anticoagulants was associated with a 22% relative reduction in intraocular bleeding compared with warfarin (risk ratio, 0.78; 95% CI, 0.61-0.99). There was no significant heterogeneity observed (I2 = 4.8%, P = .40). Comparably lower risks of intraocular bleeding with novel oral anticoagulants were seen in subgroup analyses, with no significant difference according to the indication for anticoagulation (P for heterogeneity = .49) or the novel oral anticoagulant type (P for heterogeneity = .15). Summary estimates did not differ materially when random-effects meta-analytic techniques were used. Conclusions and Relevance: These results suggest that novel oral anticoagulants reduce the risk of intraocular bleeding by approximately one-fifth compared with warfarin. Similar benefits were seen in both patients with atrial fibrillation and venous thromboembolism. Our data have particular relevance for patients at higher risk of spontaneous retinal and subretinal bleeding. These findings may also have important implications in the perioperative period, in which the use of novel oral anticoagulants may be superior. Future studies are required to better characterize the optimal management of patients with both ophthalmic disease and cardiovascular comorbidities requiring anticoagulation.

Mechanisms of neuroprotection by glucose in rat retinal cell cultures subjected to respiratory inhibition.

PURPOSE: Previous experiments have demonstrated that short-term hyperglycemia in rats renders the retina resistant to subsequent metabolic insults. The present study aimed to elucidate putative mechanisms involved in this protective response. METHODS: Retinal cultures comprising neurons and glia were treated with the mitochondrial complex I inhibitor, rotenone, at a range of concentrations, for up to 24 hours. In some cases, glucose or the alternative energy substrates, pyruvate or lactate, and/or inhibitors of glycolysis or the pentose phosphate pathway (PPP) were also applied. Cell viability was assessed using complementary techniques: immunocytochemistry, immunoblotting, cytotoxicity assay, and TUNEL. Cellular levels of ATP, reactive oxygen species (ROS), and nicotinamide adenine dinucleotide phosphate (NAD[P]H) were also assessed. RESULTS: Rotenone caused the preferential loss of neurons from retinal cultures in a concentration-dependent manner; glial cells were also affected, but only at a higher concentrations (10 µM). Cell loss was by apoptosis, and was preceded by a reduction of both cellular ATP and NAD(P)H levels and an increase in the production of ROS. Glucose counteracted the detrimental effects of rotenone. This involved a reduction in ROS levels and an increase in the cellular ATP/NAD(P)H ratio. The protective effect of glucose was partially reversed by either PPP or glycolysis inhibition. CONCLUSIONS: Glucose rescued cultured rat retinal cells from rotenone-induced toxicity. Glucose acted via both the PPP and the glycolytic pathway, maintaining cellular ATP and NAD(P)H levels and reducing ROS production. These data have implications for treatment of retinal diseases that involve metabolic compromise to neurons.

Bioenergetic-based neuroprotection and glaucoma.

Primary open-angle glaucoma (POAG) is a pressure-sensitive optic neuropathy which results in the death of retinal ganglion cells and causes associated loss of vision. Presently, the only accepted treatment strategy is to lower the intraocular pressure; however, for some patients this is insufficient to prevent progressive disease. Although the pathogenesis of POAG remains unclear, there is considerable evidence that energy failure at the optic nerve head may be involved. Neuroprotection, a strategy which directly enhances the survival of neurons, is desirable, but remains clinically elusive. One particular form of neuroprotection involves the notion of enhancing the energy supply of neurons. These 'bioenergetic' methods of neuroprotection have proven successful in animal models of other neurodegenerative diseases and conditions, including Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis and traumatic brain injury, but have been relatively unexplored in glaucoma models. This review focuses on some of the potential approaches for bioenergetic neuroprotection in the retina, including increasing the energy buffering capacity of damaged cells, decreasing the permeability of the mitochondrial membrane pore and free radical scavenging.