Intravitreal enzyme replacement for inherited retinal diseases.

PURPOSE OF REVIEW: This paper provides an update on intravitreal (IVT) enzyme replacement therapy (ERT) in metabolic retinal diseases; particularly neuronal ceroid lipofuscinosis type 2 (CLN2) also known as Batten disease. RECENT FINDINGS: ERT is being explored in CLN2 related Batten disease, a fatal neurodegenerative condition associated with retinopathy and blindness that is caused by the deficiency of lysosomal enzyme TPP1. Cerliponase alfa, a recombinant human tripeptidyl-peptidase1 (rhTPP1) administered by intraventricular infusions has been demonstrated to slow the rate of neurodegenerative decline but not retinopathy. A preclinical study of IVT rhTPP1 in a CLN2 canine model demonstrated efficacy in preserving retinal function and retinal morphology shown on histology. More recently, intravitreal (IVT) administration of rhTPP1 was reported in a first-in-human compassionate use study. Patients received 12-18 months of 8-weekly IVT ERT (0.2 mg rhTPP-1 in 0.05 ml) in one eye. No significant ocular adverse reactions were reported. Treatment decreased the rate of retinal thinning but modestly. SUMMARY: The evidence suggests that IVT ERT with rhTPP1 may be a safe and effective treatment for CLN2 retinopathy. However, the optimal dosage and frequency to achieve the best possible outcomes requires further investigation as does patient selection.

First in man study of intravitreal tripeptidyl peptidase 1 for CLN2 retinopathy.

BACKGROUND/OBJECTIVES: CLN2 Batten Disease is a fatal neurodegenerative condition of childhood associated with retinal dystrophy and blindness. Intracerebroventricular infusion of rhTPP1 greatly slows the rate of neurodegenerative decline but not retinopathy. Intravitreal rhTPP1 is known to slow retinal degeneration in a canine model of CLN2. We report a first-in-man controlled clinical trial of intravitreal rhTPP1 for CLN2 associated retinal dystrophy. SUBJECTS/METHODS: 8 children aged 5-9 with CLN2 Batten Disease were prospectively enroled. Severely affected patients were preferentially selected, provided that vision was better than no perception of light. Children underwent 8 weekly intravitreal injections of rhTPP1 (0.2 mg in 0.05 ml) into the right eye for 12-18 months. The left eye was untreated and acts as a paired control. The primary outcome was safety based on the clinical detection of complications. A secondary outcome was paracentral macular volume (PMV) measured by spectral domain OCT. Linear regression/paired t tests were used to compare rates of decline. RESULTS: No severe adverse reactions (uveitis, raised IOP, media opacity) occurred. The mean baseline PMV was 1.28 mm3(right), 1.27 mm3(left). 3 of the youngest patients exhibited bilateral progressive retinal thinning (p < 0.05), whereas retinal volume was stable in the remaining 5 patients. In the 3 patients undergoing retinal degeneration, the rate of PMV loss was slower in the treated vs. untreated eye (p = 0.000042, p = 0.0011, p = 0.00022). CONCLUSIONS: Intravitreal rhTPP1 appears to be a safe and effective treatment for CLN2 related retinopathy however commencement of treatment early in the course of disease is more likely to be efficacious.

Systemic gene therapy rescues retinal dysfunction and hearing loss in a model of Norrie disease.

Deafness affects 5% of the world's population, yet there is a lack of treatments to prevent hearing loss due to genetic causes. Norrie disease is a recessive X-linked disorder, caused by NDP gene mutation. It manifests as blindness at birth and progressive sensorineural hearing loss, leading to debilitating dual sensory deprivation. To develop a gene therapy, we used a Norrie disease mouse model (Ndptm1Wbrg ), which recapitulates abnormal retinal vascularisation and progressive hearing loss. We delivered human NDP cDNA by intravenous injection of adeno-associated viral vector (AAV)9 at neonatal, juvenile and young adult pathological stages and investigated its therapeutic effects on the retina and cochlea. Neonatal treatment prevented the death of the sensory cochlear hair cells and rescued cochlear disease biomarkers as demonstrated by RNAseq and physiological measurements of auditory function. Retinal vascularisation and electroretinograms were restored to normal by neonatal treatment. Delivery of NDP gene therapy after the onset of the degenerative inner ear disease also ameliorated the cochlear pathology, supporting the feasibility of a clinical treatment for progressive hearing loss in people with Norrie disease.

Evolution and Applications of Artificial Intelligence to Cataract Surgery.

Topic: Despite significant recent advances in artificial intelligence (AI) technology within several ophthalmic subspecialties, AI seems to be underutilized in the diagnosis and management of cataracts. In this article, we review AI technology that may soon become central to the cataract surgical pathway, from diagnosis to completion of surgery. Clinical Relevance: This review describes recent advances in AI in the preoperative, intraoperative, and postoperative phase of cataract surgery, demonstrating its impact on the pathway and the surgical team. Methods: A systematic search of PubMed was conducted to identify relevant publications on the topic of AI for cataract surgery. Articles of high quality and relevance to the topic were selected. Results: Before surgery, diagnosis and grading of cataracts through AI-based image analysis has been demonstrated in several research settings. Optimal intraocular lens (IOL) power to achieve the desired postoperative refraction can be calculated with a higher degree of accuracy using AI-based modeling compared with traditional IOL formulae. During surgery, innovative AI-based video analysis tools are in development, promoting a paradigm shift for documentation, storage, and cataloging libraries of surgical videos with applications for teaching and training, complication review, and surgical research. Situation-aware computer-assisted devices can be connected to surgical microscopes for automated video capture and cloud storage upload. Artificial intelligence-based software can provide workflow analysis, tool detection, and video segmentation for skill evaluation by the surgeon and the trainee. Mixed reality features, such as real-time intraoperative warnings, may have a role in improving surgical decision-making with the key aim of reducing complications by recognizing surgical risks in advance and alerting the operator to them. For the management of patient flow through the pathway, AI-based mathematical models generating patient referral patterns are in development, as are simulations to optimize operating room use. In the postoperative phase, AI has been shown to predict the posterior capsule status with reasonable accuracy, and can therefore improve the triage pathway in the treatment of posterior capsular opacification. Discussion: Artificial intelligence for cataract surgery will be as relevant as in other subspecialties of ophthalmology and will eventually constitute a future cornerstone for an enhanced cataract surgery pathway.

Spectrum of Mutations in NDP Resulting in Ocular Disease; a Systematic Review.

Aims and Rationale: The inner retina is supplied by three intraretinal capillary plexi whereas the outer retina is supplied by the choroidal circulation: NDP is essential for normal intraretinal vascularisation. Pathogenic variants in NDP (Xp11.3) may result in either a severe retinal phenotype associated with hearing loss (Norrie Disease) or a moderate retinal phenotype (Familial Exudative Vitreoretinopathy, FEVR). However, little is known about whether the nature or location of the NDP variant is predictive of severity. In this systematic review we summarise all reported NDP variants and draw conclusions about whether the nature of the NDP variant is predictive of the severity of the resulting ocular pathology and associated hearing loss and intellectual disability. Findings: 201 different variants in the NDP gene have been reported as disease-causing. The pathological phenotype that may result from a disease-causing NDP variant is quite diverse but generally comprises a consistent cluster of features (retinal hypovascularisation, exudation, persistent foetal vasculature, tractional/exudative retinal detachment, intellectual disability and hearing loss) that vary predictably with severity. Previous reviews have found no clear pattern in the nature of NDP mutations that cause either FEVR or Norrie disease, with the exception that mutations affecting cysteine residues have been associated with Norrie Disease and that visual loss amongst patients with Norrie disease tends to be more severe if the NDP mutation results in an early termination of translation as opposed to a missense related amino acid change. A key limitation of previous reviews has been variability in the case definition of Norrie disease and FEVR amongst authors. We thus reclassified patients into two groups based only on the severity of their retinal disease. Of the reported pathogenic variants that have been described in more than one patient, we found that any given variant caused an equivalent severity of retinopathy each time it was reported with very few exceptions. We therefore conclude that specific NDP mutations generally result in a consistent retinal phenotype each time they arise. Reports by different authors of the same variant causing either FEVR or Norrie disease conflict primarily due to variability in the authors' respective case definitions rather than true differences in disease severity.

Automated feature-based grading and progression analysis of diabetic retinopathy.

BACKGROUND: In diabetic retinopathy (DR) screening programmes feature-based grading guidelines are used by human graders. However, recent deep learning approaches have focused on end to end learning, based on labelled data at the whole image level. Most predictions from such software offer a direct grading output without information about the retinal features responsible for the grade. In this work, we demonstrate a feature based retinal image analysis system, which aims to support flexible grading and monitor progression. METHODS: The system was evaluated against images that had been graded according to two different grading systems; The International Clinical Diabetic Retinopathy and Diabetic Macular Oedema Severity Scale and the UK's National Screening Committee guidelines. RESULTS: External evaluation on large datasets collected from three nations (Kenya, Saudi Arabia and China) was carried out. On a DR referable level, sensitivity did not vary significantly between different DR grading schemes (91.2-94.2.0%) and there were excellent specificity values above 93% in all image sets. More importantly, no cases of severe non-proliferative DR, proliferative DR or DMO were missed. CONCLUSIONS: We demonstrate the potential of an AI feature-based DR grading system that is not constrained to any specific grading scheme.

Acute Angle Closure in Knobloch Syndrome.

We report cases of acute angle closure in 2 young highly myopic siblings with Knobloch syndrome. To our knowledge, this is the first report of acute angle closure in Knobloch syndrome. Both patients were homozygous for a likely pathogenic variant in COL18A1. Both responded to treatment with cyclophotocoagulation and remained stable despite declining or being medically unfit for clear lens extraction. We argue that the recent implication of heterozygous mutations in COL18A1 in familial angle closure supports the argument that acute angle closure in these 2 patients was likely to be a thus far unreported feature of Knobloch syndrome. In addition, these cases also support the hypothesis that pathogenic variants in COL18A1 may be a risk factor for acute angle closure.

Evidence Based Prediction and Progression Monitoring on Retinal Images from Three Nations.

Purpose: The aim of this work is to demonstrate how a retinal image analysis system, DAPHNE, supports the optimization of diabetic retinopathy (DR) screening programs for grading color fundus photography. Method: Retinal image sets, graded by trained and certified human graders, were acquired from Saudi Arabia, China, and Kenya. Each image was subsequently analyzed by the DAPHNE automated software. The sensitivity, specificity, and positive and negative predictive values for the detection of referable DR or diabetic macular edema were evaluated, taking human grading or clinical assessment outcomes to be the gold standard. The automated software's ability to identify co-pathology and to correctly label DR lesions was also assessed. Results: In all three datasets the agreement between the automated software and human grading was between 0.84 to 0.88. Sensitivity did not vary significantly between populations (94.28%-97.1%) with specificity ranging between 90.33% to 92.12%. There were excellent negative predictive values above 93% in all image sets. The software was able to monitor DR progression between baseline and follow-up images with the changes visualized. No cases of proliferative DR or DME were missed in the referable recommendations. Conclusions: The DAPHNE automated software demonstrated its ability not only to grade images but also to reliably monitor and visualize progression. Therefore it has the potential to assist timely image analysis in patients with diabetes in varied populations and also help to discover subtle signs of sight-threatening disease onset. Translational Relevance: This article takes research on machine vision and evaluates its readiness for clinical use.

Novel non-sense mutation in RP2 (c.843_844insT/p.Arg282fs) is associated with a severe phenotype of retinitis pigmentosa without evidence of primary retinal pigment epithelium involvement.

Retinitis pigmentosa (RP) relates to a heterogeneous group of rod-cone dystrophies of varying genetic aetiology. There is currently great interest in gene replacement therapy. Phenotyping is of particular importance because some RP genes are expressed ubiquitously and it is critically important to understand which retinal layer is primarily affected. RP2 is increasingly diagnosed in patients suffering from X-linked RP, which causes outer retinal degeneration. We present a case of a previously unreported null mutation in RP2 associated with severe RP. Loss of the retinal pigment epithelium (RPE) was noted in the central macula but not around the disc or peripherally. There was therefore no evidence of independent degeneration of the RPE. Hence despite expression in all retinal cells, RP2 deficiency does not appear to be pathogenic to the RPE. This observation may be helpful in considering the promoter and route of delivery of adeno-associated viral gene therapy vectors encoding RP2.

Report on the incidence of squamous cell carcinomas affecting the eyelids in England over a 15-year period (2000-2014).

AIMS: The authors report on trends in the incidence of squamous cell carcinoma (SCC) affecting the eyelids in England over a 15-year period and identify associations between demographic factors and SCC risk. METHODS: The National Cancer Registration and Analysis Service identified all cases of eyelid SCC in England between 2000 and 2014. The crude and age-standardised rates of eyelid SCCs in England were calculated. The association of SCC with several known demographic risk factors was then examined to assess their importance in periocular cases. RESULTS: Over the 15 years studied, there were 4022 patients in England diagnosed with a first episode of SCC affecting the eyelids. The age-standardised number of reported cases rose between 2000 and 2014 by a mean of 0.0137 cases per 100 000 population per year (equivalent to a rise in SCC incidence of approximately 2% per year). The mean age-standardised incidence rate of SCC during the study period was 0.63 cases per 100 000 population per year.Age was exponentially correlated with incidence, with an approximate doubling of the risk for every decade over the age of 60. The relative risk of eyelid SCC in men compared with women was 1.9. Social deprivation quintile by income was not found to be associated with risk of SCC. CONCLUSION: The incidence of eyelid SCC in England is rising. In addition, the age-standardised and population-standardised rate of SCC is also rising. A higher risk of SCC is strongly correlated with age and male sex but not with deprivation.

Scaling the punctum and canaliculus in patients undergoing punctoplasty surgery: A prospective cohort study.

Optical Coherence Tomography (OCT) is a safe and non-invasive method of high-resolution cross-sectional imaging of tissue microstructures using infrared radiation. This study investigates how the appearance of the punctum and proximal canaliculus differs pre- and post-operatively in patients undergoing punctoplasty surgery. Patients with symptomatic punctal stenosis warranting punctoplasty surgery were prospectively invited to enrol from a single centre. Spectral OCT images of the lower punctae were captured with a Topcon 3D OCT 2000 machine pre- and post-operatively (at their follow-up appointment). Measurements were made of the maximal punctal diameter, canalicular diameter and canalicular depth. Pre- and post-operative measurements were compared using a paired t-test. Twenty-three punctae of 18 patients with punctal stenosis were included in the study. They were 10 males and 8 females with a median age 71.0 (SD 12.1). Mean canalicular cross-sectional area was 56.9 × 10-3 mm2 pre-operatively and 267.2 × 10-3 mm2 post-operatively, showing a statistically significant increase (p = 0.0004). There was an increase in both mean canalicular width (0.253 to 0.524mm (p = 0.0001)) and depth (0.433 to 0.852mm (p = 0.0001)) from pre- to post-operatively. There was a significant improvement in symptoms from pre- to post-operatively, as measured subjectively by the Lac-Q questionnaire (p = 0.021). This study describes the change in the appearance of the punctum and proximal canaliculus in patients undergoing punctoplasty operations, by using spectral OCT to capture in vivo high-resolution images. It demonstrates that punctal OCT can be successfully applied to quantify the morphological changes of the punctum and canaliculus pre- and post-punctoplasty.

Man versus Machine: Software Training for Surgeons-An Objective Evaluation of Human and Computer-Based Training Tools for Cataract Surgical Performance.

This study aimed to address two queries: firstly, the relationship between two cataract surgical feedback tools for training, one human and one software based, and, secondly, evaluating microscope control during phacoemulsification using the software. Videos of surgeons with varying experience were enrolled and independently scored with the validated PhacoTrack motion capture software and the Objective Structured Assessment of Cataract Surgical Skill (OSACCS) human scoring tool. Microscope centration and path length travelled were also evaluated with the PhacoTrack software. Twenty-two videos correlated PhacoTrack motion capture with OSACCS. The PhacoTrack path length, number of movements, and total procedure time were found to have high levels of Spearman's rank correlation of -0.6792619 (p = 0.001), -0.6652021 (p = 0.002), and -0.771529 (p = 0001), respectively, with OSACCS. Sixty-two videos evaluated microscope camera control. Novice surgeons had their camera off the pupil centre at a far greater mean distance (SD) of 6.9 (3.3) mm, compared with experts of 3.6 (1.6) mm (p ≪ 0.05). The expert surgeons maintained good microscope camera control and limited total pupil path length travelled 2512 (1031) mm compared with novices of 4049 (2709) mm (p ≪ 0.05). Good agreement between human and machine quantified measurements of surgical skill exists. Our results demonstrate that surrogate markers for camera control are predictors of surgical skills.

Feasibility of Human Factors Immersive Simulation Training in Ophthalmology: The London Pilot.

IMPORTANCE: Human factors training can enhance teamworking and reduce error. It is used regularly in certain medical disciplines, but its use has not been established for ophthalmology to our knowledge. OBJECTIVE: To explore the feasibility of providing immersive simulation human factors training for ophthalmic surgical teams. DESIGN, SETTING, AND PARTICIPANTS: Prospective scenario-based simulation and concept description at University College London Hospital and Moorfields Eye Hospital, London, England, from December 12, 2013, to March 13, 2014. At both sites, fully immersive simulated operating theater environments were used, comprising live interactive communication with patients and theater staff, full anesthetic and operating facilities, replicated patient notes, active vital signs, and the ability to contact surgical or anesthetic teams outside of the theater via telephone. Participants were consultant (attending) and trainee ophthalmic surgeons and anesthetists, operating department assistants and practitioners, and ophthalmic nursing staff. MAIN OUTCOMES AND MEASURES: The following 4 previously validated rating tools for nontechnical skills were applied to a replicated series of scenarios based on actual patient safety incidents at Moorfields Eye Hospital and in the literature: Observational Teamwork Assessment for Surgery (OTAS), Non-Technical Skills Scale (NOTECHS), Anesthetists' Non-Technical Skills (ANTS), and Non-Technical Skills for Surgeons (NOTSS). The Pearson product moment correlation coefficient was calculated for each pair of scoring tools. Intertool and interassessor reliability was established. Interassessor consistency was compared by calculating a normalized standard deviation of scores for each tool across all assessors. RESULTS: Twenty simulation scenarios, including wrong intraocular lens implantation, wrong eye operation, wrong drug administration, and wrong patient, were provided. The intertool correlations were 0.732 (95% CI, 0.271-0.919; P = .01) for NOTECHS vs ANTS, 0.922 (95% CI, 0.814-0.968; P < .001) for NOTSS vs ANTS, 0.850 (95% CI, 0.475-0.964; P < .001) for OTAS vs ANTS, 0.812 (95% CI, 0.153-0.971; P = .03) for OTAS vs NOTECHS, 0.716 (95% CI, -0.079 to 0.955; P = .07) for OTAS vs NOTSS, and 0.516 (95% CI, -0.020 to 0.822; P = .06) for NOTECHS vs NOTSS. The normalized standard deviations of scores obtained using each tool across all assessors were 0.024 (95% CI, 0.014-0.091) for NOTSS, 0.060 (95% CI, 0.034-0.225) for OTAS, 0.068 (95% CI, 0.041-0.194) for ANTS, and 0.072 (95% CI, 0.043-0.206) for NOTECHS. CONCLUSIONS AND RELEVANCE: This study describes the feasibility of a high-fidelity immersive simulation course specifically for ophthalmic surgical teams. The ANTS and NOTSS had the highest intertool and interrater consistency, respectively. Human factors simulation in ophthalmology offers a new method of teaching team members, with the potential to reduce serious ophthalmic patient safety events. Further work will define its usefulness and practical applications.

Shunt Testing In Vivo: Observational Study of Problems with Ventricular Catheter.

Most shunt obstructions happen at the inlet of the ventricular catheter. Three hundred six infusion studies from 2007 to 2011 were classified as having a typical pattern of either proximal occlusion or patency. We describe different patterns of shunt ventricular obstruction.Solid block: Cerebrospinal fluid (CSF) aspiration was impossible. Baseline pressure was without pulse waveform (respiratory waveform may be visible). A quick increase of pressure to a level compatible with the shunt's setting was recorded in response to infusion. Distal occlusion of the shunt via transcutaneous compression resulted in a rapid increase in pressure to levels above 50 mmHg. This pattern was attributed to a solid ventricular block.Slit ventricles: At baseline, a pattern similar to that of the solid block was observed. After compression, the pressure increases, the pulse waveform appears, and the intracranial pressure is often stabilized at 25-40 mmHg. It is probable that previously slit ventricles were opened during the test.Partial block: In a partial block of the ventricular catheter by an in-growing choroid plexus, the pulse waveform at baseline was observed and CSF aspiration was possible. During infusion, the pressure increased, but the pulse amplitude disappeared. During the increase in the pressure in the shunt prechamber, the connection with the ventricles is disturbed by repositioning of the plexus.Infusion study via the shunt prechamber is able to visualize ventricular obstruction of the hydrocephalus shunt.

An Automated Detection System for Microaneurysms That Is Effective across Different Racial Groups.

Patients without diabetic retinopathy (DR) represent a large proportion of the caseload seen by the DR screening service so reliable recognition of the absence of DR in digital fundus images (DFIs) is a prime focus of automated DR screening research. We investigate the use of a novel automated DR detection algorithm to assess retinal DFIs for absence of DR. A retrospective, masked, and controlled image-based study was undertaken. 17,850 DFIs of patients from six different countries were assessed for DR by the automated system and by human graders. The system's performance was compared across DFIs from the different countries/racial groups. The sensitivities for detection of DR by the automated system were Kenya 92.8%, Botswana 90.1%, Norway 93.5%, Mongolia 91.3%, China 91.9%, and UK 90.1%. The specificities were Kenya 82.7%, Botswana 83.2%, Norway 81.3%, Mongolia 82.5%, China 83.0%, and UK 79%. There was little variability in the calculated sensitivities and specificities across the six different countries involved in the study. These data suggest the possible scalability of an automated DR detection platform that enables rapid identification of patients without DR across a wide range of races.

Optical coherence tomography imaging of the proximal lacrimal system.

INTRODUCTION: There are currently no routinely used imaging modalities for the proximal lacrimal system. Optical Coherence Tomography (OCT) is a safe and non-invasive method of high resolution cross-sectional imaging of tissue microstructures using infra-red radiation. In this study we investigate whether OCT may be used to image the punctum and proximal canaliculus. METHODS: A cohort of healthy subjects with normal ocular anatomy and no symptoms of epiphora were prospectively invited to enrol. Spectral OCT images of the lower punctae were captured with a Topcon 3D Optical Coherence Tomography 2000 machine. Measurements were made of the maximal punctal diameter, canalicular diameter and canalicular depth. Our data for depth of the vertical canaliculus was compared to the widely quoted figure of 2 mm using a two-tailed t-test to check for a statistically significant difference at p < 0.05. RESULTS: Thirty-six punctae of 18 subjects were scanned. The punctum was recognisable on the OCT image in all cases. The mean depth, width and cross- sectional area of the visualised canaliculi were 0.753 mm (SD 0.216), 0.110 mm (SD 0.067) and 9.49 × 10(-3)mm(2), respectively. The mean width of the punctum was 0.247 mm (SD 0.078). DISCUSSION: We have demonstrated the first in-vivo high resolution images of normal punctal and vertical canalicular anatomy using spectral OCT. There is currently no other practical way to accurately image punctal and proximal canalicular morphology in vivo. OCT is a convenient and readily available tool in most eye clinics with resolution ideally suited for imaging of the punctum and proximal canaliculus.

Deferred use of bevacizumab for recurrent glioblastoma is not associated with diminished efficacy.

BACKGROUND: The optimal timing to initiate bevacizumab (BV) therapy for recurrent glioblastoma (GBM) is currently unclear. To address this issue, we examined progression-free survival (PFS) and survival time (ST) in a large retrospective cohort of GBM patients treated with BV at different recurrences. METHODS: We identified 468 primary GBM patients who underwent biopsy or surgery followed by radiation therapy and temozolomide (RT/TMZ), and then received BV. PFS and ST were compared between patients stratified by the recurrence that BV was initiated (upfront, first recurrence, second recurrence, or 3+ recurrences). We also examined the effect on PFS and ST of the addition of chemotherapy to BV. In a larger cohort of GBM patients, we determined overall treatment continuation rates at each recurrence and identified variables predictive of inability to continue treatment. RESULTS: BV PFS was similar for all 3 recurrence groups (median, 4.1 months). There were no differences in BV ST (median, 9.8 months). The addition of chemotherapy to BV improved PFS but not ST. Analysis of treatment continuation rates indicated that the number of patients unable to undergo further treatments is modest, and that patients unable to tolerate BV delay can be identified by age ≥60 years and low extent of resection. CONCLUSIONS: Deferred use of bevacizumab is not associated with diminished efficacy. Analysis of treatment continuation rates identified patients who may be unable to delay BV therapy. Our findings suggest that there is a fixed survival after BV initiation and that delayed BV treatment is preferable for most patients.