Correlation of optic disc morphology and ocular perfusion parameters in patients with primary open angle glaucoma
Abstract.
Purpose: Little information is available about the relationship between glaucomatous visual field defects, morphological changes of the optic disc and ocular blood flow. In this study, ocular blood flow parameters were correlated with parameters of optic nerve head (ONH) morphology and visual field performance in a cross-sectional study.
Methods: A total of 103 patients with primary open angle glaucoma were included. Choroidal and ONH blood flow was assessed using laser Doppler flowmetry. Retinal blood velocities and retinal vessel diameters were measured with laser Doppler velocimetry and a Retinal Vessel Analyzer, respectively. To evaluate the ONH morphology, fundus photographs were taken and confocal laser scanning tomography was performed.
Results: Among all measured ocular hemodynamic parameters, the ONH blood flow was most strongly correlated to structural parameters of ONH damage and visual field loss. Reduced retinal vessel diameters were only slightly correlated with the degree of glaucomatous damage.
Conclusion: Reduced blood flow in the ONH was associated with increasing amount of visual field defect and morphological changes of the ONH. Retinal vessel diameters were only marginally associated with glaucomatous optic nerve damage. Based on retinal vessel diameter determination alone, it is not possible to assess whether reduced retinal blood flow is causative or secondary in glaucoma.
Introduction
Glaucomatous optic neuropathy is characterized by progressive loss of retinal ganglion cells including their axons and by tissue remodelling of the optic nerve head (ONH). Compromised blood supply and therefore reduced ocular blood flow (OBF) because of either increased intraocular pressure (IOP) or vascular dysregulation has been implicated in the pathogenesis of the disease.(Flammer et al. 2002; Harris et al. 1994 & 2008; Resch et al. 2009) There is likely a combination of events that sets off damage to axons or axon segments as well as support structures, and at a critical point, a cascade of events occur.
According to the vascular theory, localized damage may occur when the ocular perfusion pressure falls outside the normal range of autoregulation. This hypothesis is supported by various large-scale studies indicating that low ocular perfusion pressure is a risk factor for the prevalence, incidence and progression of glaucoma.(Bonomi et al. 2000; Leske 2007; Leske et al. 2007) In addition, nocturnal blood pressure over-dipping (Hayreh 1997; Graham & Drance 1999), as well as increased diurnal variability of ocular perfusion pressure (Choi et al. 2006) has been identified as a risk factor for glaucoma. There is evidence from several studies that patients with glaucoma show abnormal autoregulation (Grunwald et al. 1984; Fuchsjager-Mayrl et al. 2004; Galambos et al. 2006; Feke & Pasquale 2008) therefore increasing the risk of ischaemic periods.
The aim of this study was to investigate whether there is an association between visual field loss and morphological ONH damage with different parameters of OBF. This was carried out in a cross-sectional study in patients with mild to moderate degrees of glaucomatous damage.
Subjects and Methods
Subjects
The study protocol was approved by the Ethics Committee of the Medical University of Vienna. One hundred and three patients (age: 69 ± 10; 57 female, 46 male patients) with primary open angle glaucoma (OAG) were included. All subjects signed a written informed consent and passed a screening examination that included medical history and physical examination. An ophthalmic examination, including visual field testing, visual acuity with Snellen tables and applanation tonometry was performed in each patient prior to the study day.
Inclusion and exclusion criteria
Inclusion criteria were unilateral or bilateral OAG defined as pathological optic disc appearance and characteristic visual field loss (Humphrey VFA, 24-2 with near correction, SITA pac; Carl-Zeiss Meditec Co., Oberkochen, Germany), ametropia of less than three diopters, and anisometropia of less than one diopter. All patients were experienced in undergoing visual field testing, having performed at least three tests in total, one within 3 months of the start of the study. An abnormal visual field was defined as a glaucoma hemifield test outside normal limits and/or a corrected pattern standard deviation with a p-value of <0.05.(Keltner et al. 2003) Visual field eligibility criteria were <33% false-positive responses, <33% false-negative responses and <33% fixation losses. Any of the following conditions excluded a glaucoma patient from participation in the study: exfoliation glaucoma, pigmentary glaucoma, history of acute angle closure, mean deviation (MD) of the visual field of −10 decibel or worse (this limit was introduced to obtain sufficient ocular blood flow readings, because in patients with advanced disease, fixation losses make optical blood flow measurements extremely difficult), intraocular surgery or argon laser trabeculoplasty within the last 6 months, ocular inflammation or infection within the last 3 months and pregnancy. Further exclusion criteria were ocular disease other than glaucoma that might interfere with the purposes of the study, a topical or systemic therapy with steroids, and a history of an IOP >30 mmHg (untreated). Only one eye of each patient was included. If only one eye fulfilled the inclusion/exclusion criteria, this eye was chosen as study eye. In all other cases, the right eye was chosen.
Experimental design
One study day was scheduled for each subject. Dilatation of the pupil was obtained with tropicamide (Mydriaticum “Agepha”-Augentropfen, Agepha, Vienna, Austria). At the study day, measurements of ONH blood flow [Laser Doppler flowmetry (LDF)], retinal blood flow (RBF) [combining laser Doppler velocimetry (LDV) and measurements with the retinal vessel analyzer (RVA)], and of choroidal blood flow (CBF) [LDF] were carried out. Blood pressure (supine and standing) and heart rate were recorded non-invasively.
Optic nerve head morphology parameters were documented by fundus photography and scanning laser ophthalmoscopy (SLO). At the study visit, a full ophthalmological examination with IOP measurement was performed and the use of ocular as well as systemic concomitant medication was recorded. In addition, visual field testing using automated perimetry was carried out.
Patients refrained from smoking and caffeine on the study day. Prior to any measurement, care was taken to allow the subjects to maintain stable hemodynamic conditions. Therefore a 20-min resting period was scheduled before the assessment of ocular and systemic hemodynamics. Subjects remained seated between the different blood flow measurement procedures.
The study was performed at the Department of Clinical Pharmacology, Allgemeines Krankenhaus, Vienna, where IOP and OBF measurements were performed. The scanning laser tomograms and the visual field testing using automated perimetry were carried out in a cooperating ophthalmologist’s office (AH). Optic nerve head morphology parameters documented through fundus photography and SLO were evaluated by an independent site (FR and JBJ) in a masked manner. All readers were unaware of the IOP, visual field or OBF data. Ocular blood flow parameters were evaluated in Vienna, with the local investigators being not aware of the results of the morphologic assessment of the ONH.
Methods
Retinal vessel analyzer
The RVA (Imedos, Jena, Germany) is a commercially available system which comprises a fundus camera (Zeiss FF 450, Jena, Germany), a video camera, a high-resolution video recorder, a real-time monitor and a personal computer with a vessel diameter analysing software. The RVA allows the determination of retinal vessel diameter with excellent reproducibility and sensitivity.(Polak et al. 2000) In this study, retinal arteries (RAD) and veins (RVD) were studied between 1 and 2 disc diameters from the margin of the ONH.
Laser doppler velocimetry
The principle of red blood cell velocity measurement by LDV is based on the optical Doppler effect. Laser light, scattered by moving erythrocytes, is shifted in frequency. This frequency shift is proportional to the RBF velocity. The maximum Doppler shift corresponds to the centreline erythrocyte velocity (Vmax).(Riva et al. 1985) Mean RBF velocity was calculated as Vmean = Vmax/2. In this study, we used a fundus camera-based system (Oculix 4000; Oculix Sarl, Arbaz, Switzerland). Retinal blood flow velocity was measured in the temporal inferior veins only at the same locations as vessel diameters of the temporal inferior veins were measured using the RVA.
Laser doppler flowmetry
Choroidal and ONH blood flow were assessed with a fundus camera–based laser Doppler flowmeter (Oculix 4000) (Bonner & Nossal 1990; Riva et al. 1992, 1994) With this technique, the vascularized tissue is illuminated by laser light. Scattering on moving blood cells (RBCs) leads to a frequency shift in the backscattered light, from which the mean RBC velocity, the blood volume and the blood flow can be calculated in relative units. The laser beam was directed to the fovea to assess CBF and on the temporal neuroretinal rim to assess ONH blood flow.
Blood pressure measurement
Systemic blood pressures were measured at the upper arm by an automated oscillometric device (HP-CMS patient monitor, Hewlett Packard GmbH, Palo Alto, CA, USA). Pulse rate was automatically recorded from a finger pulse-oxymetric device. Ocular perfusion pressure (OPP) in the sitting position was calculated as 2/3 MAP – IOP.
Stereo fundus photography
For all eyes, 15° colour stereo optic disc transparencies were taken with a telecentric 30° fundus camera equipped with a 15° converter (Zeiss FF 450). The disc slides were projected in a scale of 1–15. The outlines of the optic cup, optic disc, and the peripapillary scleral ring were plotted on paper and morphometrically analysed. To obtain values in absolute size units, the ocular and photographic magnifications were corrected using the Littmann method.(Littmann 1982) The optic cup was defined on the basis of contour and not of pallor. The border of the optic disc was identical with the inner side of the peripapillary scleral ring. Horizontal and vertical disc diameter and horizontal and vertical cup/disc (C/D) ratios were chosen as outcome parameters. (Drance 1998; Jonas et al. 2002)
Confocal SLO
The Heidelberg Retina Tomograph (HRT II; Heidelberg Engineering, Dossenheim, Germany) employs confocal scanning technology to provide topographical measures of the ONH and parapapillary retina.( Weinreb et al. 1993; Burk et al. 2000) Three 15° field-of-view scans, centred on the ONH were obtained for each tested eye. A mean topography image of these three scans was created using HRT II software. The ONH margin was outlined on the mean image using information obtained by viewing simultaneous stereoscopic photographs of the ONH. The C/D area ratio and the retinal nerve fibre layer (RNFL) cross-sectional area were chosen as outcome variables.
Automated visual field testing
Visual field testing was performed with the Humphrey Field analyzer (Humphrey VFA, 24-2 with near correction, SITA pac, Munich, Germany). All patients were experienced in visual field testing, having performed at least three tests in total, one within 3 months of the beginning of the study. Visual field eligibility criteria were <33% false-positive responses, <33% false-negative responses and <33% fixation losses.
Measurement of intraocular pressure with Goldmann applanation tonometry
The IOP was measured with an applanation tonometer integrated into the slitlamp before pupil dilatation.
Statistics
The correlation between ocular hemodynamic and parameters of ONH damage and visual field MD on the other hand were analysed with linear regression analysis. In addition, the association between ocular hemodynamic variables and OPP as well as the association between ocular hemodynamic variables and topical antiglaucoma medication was tested. To evaluate the significance of associations, a multiple regression model was employed. For this multiple regression analysis, the ocular blood flow variables were chosen as independent variables and the measures of glaucomatous damage (MD, Horizontal disc diameter, Vertical disc diameter, etc.) were chosen as independent variables. A forward stepwise approach was chosen in this multiple regression model. A similar procedure was applied to investigate whether parameters of visual field damage were associated with parameters of the ONH morphology. A two-tailed p-value of <0.05 was considered the level of significance. All statistical analysis was performed with the Statistica® software package (Release 6.0, StatSoft Inc., Tulsa, OK, USA).
Results
Patient characteristics, systemic hemodynamic measurements, IOP, visual field MD and OBF parameters are given in Table 1. Because of insufficient readings, subfoveal CBF data for one patient had to be excluded. All other variables stem from 103 subjects. Most of the glaucoma patients were under IOP lowering therapy (Table 2).
| Patients with glaucoma (n = 103) | |
|---|---|
| Sex (male/female) | 46/57 |
| Age (years) | 68.8 (42–87) |
| Intraocular pressure (mmHg) | 16.2 (10–27) |
| Mean arterial pressure (mmHg) | 92.6 (67–151) |
| Ocular perfusion pressure (mmHg) | 44.2 (30–79) |
| Pulse rate (bpm) | 67.3 (47–95) |
| Retinal blood velocity (Temporal inferior vein, cm/s) | 2.05 (1.00–3.60) |
| Arterial diameter (Temporal inferior artery, μm) | 122 (83–171) |
| Arterial diameter (Temporal superior artery, μm) | 115 (65–147) |
| Arterial diameter (Nasal inferior artery, μm) | 108 (76–134) |
| Arterial diameter (Nasal superior artery, μm) | 106 (80–134) |
| Venous diameter (Temporal inferior vein, μm) | 143 (97–199) |
| Venous diameter (Temporal superior vein, μm) | 138 (100–183) |
| Venous diameter (Nasal inferior vein, μm) | 117 (81–156) |
| Venous diameter (Nasal superior vein, μm) | 117 (83–154) |
| Optic nerve head blood flow (a.u.) | 12.20 (6.50–18.12) |
| Choroidal blood flow (a.u.) | 11.35 (4.40–19.50) |
| Horizontal disc diameter (fundus photo, mm) | 1.25 (0.62–2.34) |
| Vertical disc diameter (fundus photo, mm) | 1.30 (0.68–2.34) |
| Horizontal cup/disc ratio (fundus photo) | 0.34 (0.05–0.80) |
| Vertical cup/disc ratio (fundus photo) | 0.40 (0.10–0.87) |
| Disc area (HRT, mm2) | 1.81 (0.73–2.59) |
| Cup/disc area ratio (HRT) | 0.41 (0.03–0.75) |
| RNFL cross-sectional area (mm2) | 0.17 (−0.07–0.41) |
| MD (db) | −2.40 (−9.70–2.90) |
- MD = mean deviation, RNFL = retinal nerve fibre layer.
| Glaucoma medication | Number of patients under medication |
|---|---|
| Latanoprost | 45 |
| Brinzolamide | 11 |
| Timolol | 9 |
| Timolol + Dorzolamide (fixed combination) | 24 |
| Travoprost | 4 |
| Travoprost + Timolol | 1 |
| Bimatoprost | 14 |
| Brimonidine | 8 |
| Acetazolamide systemically | 4 |
| Timolol gel-forming solution | 2 |
| Betaxolol | 2 |
| Travoprost + Timolol (fixed combination) | 1 |
| Bimatoprost + Timolol (fixed combination) | 1 |
| Latanoprost + Timolol (fixed combination) | 2 |
| Nimodipine systemically | 1 |
Fundus photographs with sufficient quality for evaluation were obtained from 92 of 103 patients. The mean optic disc diameters, both horizontally and vertically, were within the expected range. The horizontal disc diameter ranged between 0.62 and 2.34 mm. The vertical disc diameter varied between 0.68 and 2.34 mm (Table 1). In 85 subjects, HRT images of sufficient image quality were obtained.
Linear correlation analysis revealed that the worse the MD the larger was the vertical C/D ratio (r = 0.38, p = 0.001) and the horizontal C/D ratio (r = 0.33, p = 0.004). The worse the MD the higher was the C/D area ratio (r = −0.27; p = 0.032) and the lower was the RNFL cross-sectional area (r = 0.29; p = 0.022). In addition, worse MD was associated with higher IOPs (r = −0.28, p = 0.013) and lower MAPs (r = 0.24, p = 0.039).
Linear correlation analysis also showed that the ONH blood flow was positively correlated with the MD (r = 0.45, p < 0.001), and it was negatively correlated with the horizontal C/D ratio (r = −0.43, p < 0.001) and the vertical C/D ratio (r = −0.35, p = 0.002). By contrast, the CBF was not significantly associated with the MD or the morphologic parameters of the ONH. A significant and positive correlation was detected between the RBF velocity and the MD (r = 0.35, p = 0.002). The RBF velocities were not correlated with ONH morphology. The association between the RADs and the measures of glaucomatous damage was generally weak. A significant correlation was only found between the retinal temporal inferior diameter and the horizontal C/D ratio (r = −0.26, p = 0.026), the temporal inferior diameter and the vertical C/D ratio (r = −0.29, p = 0.011), the temporal inferior diameter and the MD (r = 0.24, p = 0.035), the nasal inferior diameter and the horizontal C/D ratio (r = −0.24, p = 0.039), the nasal inferior diameter and the vertical C/D ratio (r = −0.28, p = 0.014), the nasal superior diameter and the vertical C/D ratio (r = −0.25, p = 0.026), and the nasal superior diameter and the mean visual field loss (r = 0.27, p = 0.021). Correlations between RVDs and parameters of glaucomatous damage were only significant for the nasal inferior vein with the horizontal C/D ratio (r = −0.28, p = 0.013) and with the vertical C/D ratio (r = −0.32, p = 0.004). Neither MAP nor IOP were associated with any of the morphological parameters.
Only few significant correlations were found between parameters of OBF and HRT variables. Optic nerve head blood flow was negatively correlated with C/D area ratio (r = −0.29; p = 0.022) and positively correlated with the RNFL cross-sectional area (r = 0.28; p = 0.030). The RBF velocity was negatively associated with the C/D area ratio (r = −0.33; p = 0.005). All other parameters were not significantly associated.
Results of the multiregression analysis are shown in Table 3. Only few correlations turned out to be significant in this model. The ONH blood flow was associated with parameters obtained from fundus photography and HRT.
| MD | HDD | HCD | VDD | VCD | DA | CDA | RCSA | |
|---|---|---|---|---|---|---|---|---|
| Retinal blood velocity (Temporal inferior vein, cm/s) | 0.01 | 0.21 | 0.13 | 0.65 | 0.11 | 0.23 | 0.42 | 0.21 |
| Arterial diameter (Temporal inferior artery, μm) | 0.04 | 0.91 | 0.07 | 0.76 | 0.06 | 0.21 | 0.13 | 0.12 |
| Arterial diameter (Temporal superior artery, μm) | 0.17 | 0.90 | 0.09 | 0.82 | 0.67 | 0.87 | 0.11 | 0.13 |
| Arterial diameter (Nasal inferior artery, μm) | 0.12 | 0.86 | 0.07 | 0.54 | 0.09 | 0.76 | 0.19 | 0.23 |
| Arterial diameter (Nasal superior artery, μm) | 0.07 | 0.76 | 0.07 | 0.66 | 0.14 | 0.54 | 0.21 | 0.16 |
| Venous diameter (Temporal inferior vein, μm) | 0.11 | 0.23 | 0.12 | 0.65 | 0.14 | 0.76 | 0.17 | 0.21 |
| Venous diameter (Temporal superior vein, μm) | 0.12 | 0.44 | 0.31 | 0.77 | 0.19 | 0.54 | 0.46 | 0.50 |
| Venous diameter (Nasal inferior vein, μm) | 0.08 | 0.53 | 0.04 | 0.63 | 0.09 | 0.87 | 0.65 | 0.66 |
| Venous diameter (Nasal superior vein, μm) | 0.31 | 0.66 | 0.13 | 0.77 | 0.21 | 0.79 | 0.52 | 0.87 |
| Optic nerve head blood flow (a.u.) | <0.01 | 0.54 | 0.01 | 0.65 | 0.01 | 0.76 | 0.02 | 0.02 |
| Choroidal blood flow (a.u.) | 0.06 | 0.98 | 0.09 | 0.54 | 0.07 | 0.90 | 0.09 | 0.08 |
| Intraocular pressure (mmHg) | 0.04 | 0.31 | 0.24 | 0.41 | 0.21 | 0.13 | 0.24 | 0.38 |
| Mean arterial pressure (mmHg) | 0.11 | 0.67 | 0.51 | 0.46 | 0.61 | 0.34 | 0.40 | 0.31 |
- The p values are presented. MD = Mean deviation (Visual field testing), HDD = Horizontal disc diameter (fundus photo, mm), VDD = Vertical disc diameter (fundus photo, mm), HCD = Horizontal cup/disc ratio (fundus photo), VCD = Vertical cup/disc ratio (fundus photo), DA = Disc area (HRT, mm2), CDA = Cup/disc area ratio (HRT), RCSA = RNFL cross-sectional area (HRT, mm2).
None of the ocular hemodynamic parameters was associated with topical anti-glaucoma drugs used or IOP. No association was found between retinal vessel diameters and OPP. Positive correlations were, however, found between ONH blood flow and OPP (r = 0.35, p = 0.002), CBF and OPP (r = 0.36, p = 0.001) and RBF velocity and OPP (r = 0.29, p = 0.010).
Discussion
Glaucoma remains a multi-factorial optic neuropathy of unknown aetiology. Elevated IOP is the most important risk factor for the disease, although the exact pathways that relate ocular hypertension to glaucomatous optic neuropathy have not been elucidated. Mechanisms other than elevated IOP do, however, contribute to glaucomatous damage. The results of the Collaborative NTG study show for instance that IOP reduction in patients with normal-tension glaucoma does not fully prevent progression of the disease (Drance 1998).
In the present study, blood flow in the ONH as assessed with LDF was associated with the degree of visual field damage and the morphological ONH changes. By contrast, subfoveal CBF was not correlated with functional or morphological parameters. As retinal vessel diameters were only slightly associated with glaucomatous damage in the present study, it appears that this parameter is not an adequate measure to study the relation between perfusion abnormalities and glaucoma. A number of previous studies have shown that RADs are reduced in patients with glaucoma as compared with healthy control subjects which is in good agreement with the present data (Jonas et al. 1991; Mitchell et al. 2005; Wang et al. 2007). On the other hand, the Rotterdam study did not identify reduced RADs as a risk factor for incident OAG (Ikram et al. 2005). In addition, there was no difference in RADs between patients with progressive and non-progressive glaucoma (Soares et al. 2003) and retinal vessel diameters were not associated with glaucoma progression (Jonas et al. 2004). It needs to be considered that glaucoma may primarily affect the small retinal arterioles, which contribute most to vascular resistance. This is also supported by the fact that retinal blood velocity was associated with MD in our study population.
Mean blood pressure tended to be negatively associated with C/D ratios and was positively associated with MD. This finding is in agreement with the previous notion that low ocular perfusion pressure is a risk factor for the prevalence, incidence and progression of glaucoma (Bonomi et al. 2000; Leske 2007; Leske et al. 2007). The present cross-sectional study showed a correlation between impaired OBF and the stage of visual field damage and ONH damage. Whether these changes in OBF are a cause or a consequence of glaucomatous optic atrophy is still unknown. Some small-scale studies suggest that reduced OBF is a risk factor for the progression of the disease. A number of studies indicate that reduced retrobulbar flow velocities and higher resistive indices as assessed with colour Doppler imaging are associated with glaucoma progression (Galassi et al. 2003; Satilmis et al. 2003; Martinez & Sanchez 2005; Galambos et al. 2006; Zeitz et al. 2006). Reduced blood volume assessed by LDF was also associated with progression of visual field damage (Zink et al. 2003).
Unfortunately there is currently no Gold Standard technique for the measurement of ocular blood flow available. For a detailed discussion on this topic, the reader is referred to some recent review articles (Rechtman et al. 2003; Schmetterer & Garhofer 2007; Harris et al. 2008; Garhofer et al. 2010). It needs, however, to be mentioned that the scattering volume with all LDF techniques used in the present study is unknown. Hence, the sampling depth and the exact relative contributions from different tissues with this technique is unknown (Riva et al. 2010). With LDF measurements taken at the ONH, it also has to be taken into account that the disease-related loss of neuroretinal rim tissue may in itself reduce the ONH readings owing to capillary loss.
Nevertheless, a large-scale multicenter longitudinal trial investigating the relation between reduced OBF parameters and glaucoma progression is urgently required. The data of the present study indicate that relying solely on measurements of retinal vessel diameters in such a trial would not be sufficient. As measurement of retinal blood velocity using LDV is time consuming, it would make sense to include LDF technology for measuring ONH blood flow. The present study used single point LDF, but scanning LDF may represent an alternative.
In conclusion, the present cross-sectional study indicates that ONH blood flow as assessed with LDF is associated with glaucomatous damage, whereas retinal vessel diameters are only slightly associated with the disease severity. Hence, longitudinal trials on the relation of retinal and/or optic nerve head blood flow and the incidence or progression of glaucoma should not solely include measurements of retinal vessel diameters.
Acknowledgements
This study was supported by an unrestricted research grant from MSD.
