Clinical science Baseline retrobulbar blood flow is associated with both functional and structural glaucomatous progression after 4 years Nicholas A Moore, Alon Harris, Scott Wentz, Alice Chandra Verticchio Vercellin, Priyanka Parekh, Joshua Gross, Rehan M Hussain, Claudia Thieme, Brent Siesky Department of Ophthalmology, ABSTRACT factors in the disease onset and progression.56This Indiana University School of Background/aims Previous studies suggest that has led to the evaluation of ocular blood flow and Medicine, Indianapolis, Indiana, USA vascular abnormalities are involved in the pathogenesis its role in glaucoma. of open-angle glaucoma. This study aims to examine the Ischaemia to the optic nerve and reduced ocular Correspondence to relationship of baseline retrobulbar blood flow blood flow has been shown to contribute to the Professor Alon Harris, measurements with functional and structural pathogenesis and progression of glaucoma.78 Department of Ophthalmology, glaucomatous progression in patients with open-angle Several systemic and localised vascular abnormal- Indiana University School of Medicine, 1160 West Michigan glaucoma over 4 years. ities have been linked to OAG including arterial Street, Indianapolis, IN 46202, Methods In this study, 112 patients with open-angle hypertension, nocturnal hypotension, optic disc USA; [email protected] glaucoma were examined at baseline and 78 with haemorrhage, migraines and age-related changes of – retrobulbar blood flow assessments were followed to the vasculature.9 12 In addition, a decreased ocular Received 27 January 2016 4 years. Colour Doppler imaging was used to evaluate perfusion pressure has been linked to an increase in Revised 17 May 2016 13–17 Accepted 25 May 2016 retrobulbar blood flow. Structural disease progression OAG prevalence, incidence and progression. Published Online First was examined with optical coherence tomography and Patients with OAG have been found to have 13 June 2016 Heidelberg Retinal Tomography III. Functional disease reduced blood flow in the retinal, choroidal and progression was monitored with automated perimetry retrobulbar circulations which is associated with using Humphrey visual fields. Mixed-model analysis of glaucomatous functional and structural disease pro- – covariance was used to test for significance of changes gression.18 23 However, the results from these from baseline to 4-year follow-up. Two-sample t tests studies are varied, and the exact relationship and χ2 tests were used to test for baseline blood flow between ocular blood flow and glaucoma progres- differences between patients who progressed and those sion has not been fully delineated. who did not progress. Many of these previous studies are limited by Results Patients who progressed structurally had a small sample sizes, often involve a heterogeneous statistically significant lower baseline mean ophthalmic group of patients including glaucoma suspects and artery peak systolic velocity (PSV) (p=0.024) and do not evaluate both functional and structural glau- ophthalmic artery end diastolic velocity (EDV) (p=0.012) coma progression in parallel. We previously compared with those who did not progress. Similarly, a reported findings detailing the association between lower baseline mean ophthalmic artery PSV (p=0.031) reductions in retrobulbar and retinal blood flow and ophthalmic artery EDV (p=0.005) were associated over time with structural glaucomatous progres- with patients who progressed functionally compared sion.24 Here, we present data from a large 4-year with those who did not progress after 4 years. longitudinal study detailing the relationship Conclusions In this study population, lower baseline between retrobulbar blood flow and simultaneous ophthalmic artery blood flow velocities were associated structural and functional glaucoma progression. with simultaneous structural and functional glaucoma progression after 4 years. MATERIAL AND METHODS A cohort of patients with OAG were enrolled in INTRODUCTION this study and examined at baseline and every Primary open-angle glaucoma (OAG) is a multifac- 6 months for 4 years at the Glaucoma and torial optic neuropathy characterised by progressive Diagnostic Center at Indiana University School of retinal ganglion cell death and a characteristic Medicine. All patients signed an informed consent visual field loss. Elevated intraocular pressure (IOP) prior to initiation of this study, which adhered to has been identified as a major risk factor for OAG the tenets of the Declaration of Helsinki. The study and current treatments focus on reducing and con- protocol was approved by the Institutional Review trolling the IOP to limit disease progression. Board committee at the Indiana University School Despite advances in pharmacological and surgical of Medicine. interventions, however, it is well established that All participants were required to meet the fol- glaucoma progression is still observed in some lowing inclusion criteria: age 30 years or older, patients with IOP reduction. Additionally, a high confirmed OAG in at least one eye by a glaucoma percentage of individuals with elevated IOP do not specialist, best-corrected visual acuity of 20/60 or To cite: Moore NA, – Harris A, Wentz S, et al. Br J develop glaucoma.1 4 Each of these findings sug- better in the study eye, and acceptable reliability Ophthalmol 2017;101:305– gests that glaucoma progression is multifactorial in indices in previous Humphrey visual field 308. origin and there are other underlying contributing examinations. Moore NA, et al. Br J Ophthalmol 2017;101:305–308. doi:10.1136/bjophthalmol-2016-308460 305 Clinical science Patients were excluded for the following reasons: Humphrey Peripapillary retinal nerve fibre layer (RNFL) thickness and visual field damage consistent with a mean deviation (MD) less optic nerve head (ONH) parameters were assessed by optical than −15 decibels or a clinically determined threat to fixation in coherence tomography (OCT) (Stratus software V.4.0, Zeiss both hemifields, evidence of pseudoexfoliation or pigment dis- Meditec, Dublin, California, USA). The examination was per- persion, history of acute angle-closure glaucoma or a narrow formed and repeated until good-quality analysis (signal strength occludable anterior chamber angle, history of chronic or recur- <7) was obtained. Measurements were made along a circle con- rent inflammatory eye diseases, history of intraocular trauma, centric with the optical disc (Fast RNFL thickness acquisition severe or progressive retinal disease, any abnormality preventing protocol) to assess RNFL thickness. The mean RNFL thickness reliable applanation tonometry, cataract surgery within the past and cup/disc vertical and horizontal ratios were calculated by year, resting pulse <50 bpm, or severe cardiovascular, renal or OCT using the existing software. Topographic analysis of the pulmonary disease. ONH was performed using Heidelberg Retinal Tomography III One qualified eye was randomly assigned as the observational (HRT-III) (Heidelberg Engineering, Heidelberg, Germany). The study eye for each subject. All patients were questioned for their HRT-III was a secondary modality to OCT and was used to sup- clinical history, assessed for IOP by Goldmann applanation tono- plement evaluation for subtle RNFL and ONH changes. metry, and evaluated for best-corrected visual acuity. To limit Structural progression was defined as two consecutive visits with reproducibility bias with imaging, a single experienced operator an RNFL thickness decrease ≥8% and/or horizontal or vertical with over 10 years of experience performed all measurements in cup/disc ratio increase by ≥0.2 from baseline. the same order and at the same time of the day for each patient. Statistical analysis involved performing a mixed-model ana- Retrobulbar blood flow velocities and vascular resistance were lysis of covariance to test for significant change from baseline to measured with the Philips HDI 5000 colour Doppler imaging 4-year follow-up. Two-sample t tests and χ2 tests were used to (CDI) system with the microvascular small parts clinical option analyse differences in baseline data between patients who pro- using a 7.5 MHz linear probe (Philips Ultrasound, Bothell, gressed and those who did not progress. Washington, USA). Peak systolic velocity (PSV) and end diastolic velocity (EDV) were determined for the ophthalmic artery RESULTS (OA), central retinal artery (CRA), nasal posterior ciliary artery In this study, 112 patients with OAG were enrolled according to (NPCA) and temporal posterior ciliary artery (TPCA). The the prior listed inclusion and exclusion criteria and 78 patients Pourcelot’s resistive index (RI) was calculated (RI=(PSV−EDV)/ with retrobulbar blood flow assessments using CDI were fol- PSV)) for each vessel. Retrobulbar blood flow velocities obtained lowed through to 4 years. Overall baseline characteristics of the by CDI were standardised by using a printout at each visit to population revealed a mean age 64.9±11.0 years; female ensure that velocities were taken from nearly the same location (n=68), male (n=44); African descent (n=29), European in each vessel. CDI was selected as the imaging modality of descent (n=83); non-insulin-dependent diabetes mellitus choice because of its known reproducibility at measuring retro- (n=21), no diabetes mellitus (n=91). bulbar blood flow velocities and calculating the RI.25 In this study population, the RI significantly increased in all Visual function was assessed by automated perimetry with the four retrobulbar