Article Text

Download PDFPDF

Nitric oxide proxies and ocular perfusion pressure in primary open angle glaucoma
  1. F Galassi1,
  2. G Renieri1,
  3. A Sodi1,
  4. F Ucci1,
  5. L Vannozzi1,
  6. E Masini2
  1. 1Department of Oto-Neuro-Ophthalmological Sciences, University of Florence, Eye Unit II, Florence, Italy
  2. 2Department of Clinical and Preclinical Pharmacology, Florence, Italy
  1. Correspondence to: Professor F Galassi Department of Oto-Neuro-Ophthalmological Sciences, University of Florence, Eye Unit II, Via S Marta 24, 50139 Florence, Italy;


Background: To investigate the levels of nitric oxide (NO) markers in plasma and aqueous humour of patients with primary open angle glaucoma (POAG) and their relation to ocular perfusion pressure.

Methods: Cyclic guanosine monophosphate (cGMP) and nitrite (NO2) were determined in plasma and aqueous humour of 38 patients with POAG and 46 controls. Blood pressure and IOP were measured to calculate ocular perfusion pressure (PP).

Results: cGMP and NO2 plasma levels were significantly decreased in glaucoma patients compared with controls (p = 0.001 v p = 0.004). In the aqueous humour of subjects with POAG, cGMP and NO2 concentrations were also lower than in normal eyes (p = 0.0001 v p = 0.001). There was a linear association between cGMP in plasma and aqueous humour in glaucomas and controls (r = 0.514, p = 0.029 and r = 0.558, p = 0.004) and this relation differed in the two groups (p = 0.003). Considering glaucoma patients with controls, a positive correlation was found between cGMP and PP (r = 0.379, p = 0.01) and between NO2 and PP (r = 0.339, p = 0.040). The cGMP/PP correlation was of borderline statistical significance in controls (p = 0.050), whereas it did not attain statistical significance in POAG, as well as the association between NO2 and PP when glaucomas and controls were considered separately.

Conclusions: The authors found alterations of NO markers in the plasma and aqueous humour of glaucoma patients. Primary or secondary impaired NO balance could alter ocular perfusion pressure.

  • primary open angle glaucoma
  • nitric oxide
  • ocular perfusion pressure
  • cGMP, cyclic guanosine monophosphate
  • NO, nitric oxide
  • NOS, nitric oxide synthase
  • POAG, primary open angle glaucoma
  • PP, perfusion pressure
  • primary open angle glaucoma
  • nitric oxide
  • ocular perfusion pressure
  • cGMP, cyclic guanosine monophosphate
  • NO, nitric oxide
  • NOS, nitric oxide synthase
  • POAG, primary open angle glaucoma
  • PP, perfusion pressure

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Nitric oxide (NO) is synthesised in endothelial cells by type III isoform of nitric oxide synthase (NOS) and, under basal conditions of flow, it sustains a constant vasodilator tone in the vascular system. Changes in the environment, such as increased stress or hypoxia, can stimulate NO synthesis and the local regulation of vascular tone and blood flow.1

Abundant NOS immunoreactivity has been found in the optic nerve head (ONH) vasculature.2,3 At this level an increase of NOS in the vascular endothelium has been hypothesised to be neuroprotective by promoting vasodilation and tissue blood flow.3–7

Nitric oxide synthase inhibition has been shown to strongly reduce choroidal and ONH blood flow in animals8–12 and humans.13 Intravenous infusion of NG-nitro-L-arginine (L-NMMA), a NOS inhibitor, provoked a reduction of ONH blood flow in healthy subjects.13 Another work reported that the NO donor 5-isosorbide mononitrate significantly increases ONH blood flow in normal subjects mainly causing an increment in blood volume,14 and an elective sensitivity of the ONH circulation to nitrates has been postulated.15

Tissue ischaemia has been involved in the pathogenesis of glaucomatous optic neuropathy and an altered circulation has been indicated in some glaucoma patients.16–21

Plasma and aqueous humour levels of cGMP, an indirect indicator of NO, were decreased in patients with normal tension glaucoma. The same glaucomatous group showed lower systolic and diastolic velocities of the ophthalmic artery when examined with colour Doppler imaging.22 In POAG, abnormal IOP and vascular dysregulation might together determine optic nerve damage, both reducing ocular perfusion.

As NO can modulate IOP and ocular circulation,23,24 and contrasting data have been published recently on NO production25,26 in glaucoma patients, we wanted to elucidate whether NO formation is disturbed in hypertensive POAG and whether it influences ocular perfusion pressure. We therefore investigated the levels of NO2, the stable end product of NO metabolism, and of cGMP, the intracellular mediator of NO action, in aqueous humour and plasma of patients with POAG and controls.


Study population

Thirty eight POAG patients were enrolled in the study together with 46 control subjects referred to our clinic for cataract surgery. Patients and controls were matched for age, sex, and blood pressure (table 1). The following exclusion criteria were used: any ocular disease except senile cataract or POAG; diabetes; infections; cardiovascular pathologies including hypertension and atherosclerotic lesions determining haemodynamically significant stenosis; assumption of non-steroidal anti-inflammatory drugs; steroids; ACE inhibitors, or NO donors. All subjects had to follow a nitrate free diet for three days before blood extraction.

Table 1

Demographic characteristics of patients enrolled in the study

Glaucoma patients had IOP higher than 21 mm Hg without antiglaucomatous drug, open anterior chamber angle, glaucomatous disc damage (cup:disc ratio ⩾0.6), and glaucomatous field loss on the Humphrey perimeter, 30–2 full threshold programme (Humphrey Instruments, San Leandro, CA, USA). Visual field defects were included in the stage I–III, according to Aulhorn’s classification modified by Greve.27 All control subjects had a normal ocular examination, IOP within normal limits, and intact visual fields.

Written informed consent was obtained from each subject and the tenets of the Declaration of Helsinki were observed.

Measurement of cGMP and NO2 in the plasma and in the aqueous humour

Blood samples of 3 ml were collected from a forearm vein of every patient and immediately subjected to plasma separation and stored at −20°C.

Glaucoma patients and controls admitted to our clinic for cataract surgery, who met the aforementioned inclusion criteria, were also recruited for cGMP and NO2 determination in the aqueous humour. Preoperatively, the same antibiotic and the same standard mydriatic protocol was used (three or four drops of an association of tropicamide 1% and phenylephrine 2.5%) for glaucomas and controls, so the two subgroups differed only for the antiglaucomatous treatment (timolol 0.50% twice a day). Cataract extraction was performed with a phacoemulsification technique. Before performing the clear corneal tunnel, 200–400 μl of aqueous humour was quickly withdrawn by means of a paracentesis, and replaced by a viscoelastic substance. The paracentesis were always performed by the same surgeon with a tuberculine needle on a 1 ml syringe. Ten μl of a solution of 10−5 M of isobutyl methylxantine (IBMX), a phosphodiesterase inhibitor, was added to each sample. All samples were stored at −20°C.

NO levels were evaluated in plasma and aqueous humour by measuring the concentration of NO2 and of cGMP.

NO2 was measured spectrophotometrically by the Griess reaction;28 values were obtained by comparison with standard concentrations of sodium nitrite and expressed as micromoles of NO2 per milligram of protein.

cGMP levels were measured in plasma and aqueous humour after the extraction with 10% trichloroacetic acid with 0.5 M tri-n-octylamine dissolved in 1,1,2-trichlorotrifluorethane.28 cGMP was determined with the use of a radioimmunoassay kit (Amersham, Buckinghamshire, UK) after acetylation of the samples with acetic anhydride. Values are expressed as picomoles and as femtomoles of cGMP per milligram of protein for plasma and aqueous humour, respectively.

Protein concentrations were determined according to Lowry.29 All experiments were performed in duplicates by a blind operator.

Ocular perfusion pressure

Immediately before collecting the blood sample, the blood pressure was measured by sphyngmomanometry. Although patients with a previous medical diagnosis of arterial hypertension were not enrolled at all in the study, we also excluded for ocular perfusion pressure evaluation subjects who had arterial pressure higher than 145/90 mm Hg before plasma withdrawal.

Intraocular pressure was assessed by Goldmann applanation tonometry (Haag Streit, Bern, Switzerland). Mean brachial artery blood pressure (BPm) was calculated according to the formula:


and ocular perfusion pressure (PP) was derived using the following relation:


In control subjects and in patients with bilateral glaucoma the eye considered for PP calculation was chosen by random.

Statistical analysis

Results are shown as mean (standard deviation, SD). Statistical comparisons were conducted using the Student’s t test for unpaired observations. Linear regression and correlation were used to analyse the association between cGMP plasma and aqueous humour levels and between cGMP, NO2 plasma concentration and PP. Comparisons between regression lines of glaucomas and controls were performed with the overall test for coincidence. A p value of ⩽0.05 was considered significant.


cGMP and NO2 plasma levels

Forty six controls and 38 POAG subjects were eligible for cGMP assessment and among them, respectively, 19 and 23 also underwent NO2 determination. cGMP plasma concentration was significantly lower in glaucoma patients than in normals (p = 0.001, fig 1).

Figure 1

cGMP plasma levels in control (2059.68 (697.86) pmoles/mg protein) and glaucoma subjects (1568.62 (625.55) pmoles/mg protein). Values are mean (SD); p = 0.001 by the Student’s t test for unpaired data.

Also NO2 plasma levels were decreased in POAG patients compared with controls (p = 0.004, fig 2).

Figure 2

Nitrite (NO2) plasma concentrations in glaucomas (166.68 (38.45) μmol/mg protein) and controls (228.39 (76.22) μmoles/mg protein). Values are mean (SD); p = 0.004 by the Student’s t test for unpaired data.

cGMP and NO2 levels in the aqueous humour

Aqueous humour samples were collected from 18 POAG patients and 25 subjects with cataract only. The glaucoma patients had a good tonometric control with therapy. Preoperative IOP statistically differed in the two groups (16 (4.33) v 13 (3.15) mm Hg; p = 0.012).

In glaucomatous eyes, cGMP mean concentration was significantly lower than in control eyes (p = 0.0001, fig 3).

Figure 3

Amount of cGMP in the aqueous humour of patients with POAG and cataract (1559.32 (619.12) fmol/mg protein) v patients with cataract only (controls) (2931.07 (1387.01) fmol/mg protein). Student’s t test for unpaired data: p = 0.0001.

NO2 amount was determined in the aqueous humour every time we could safely withdraw a sufficient volume from the anterior chamber of the individual eye. NO2 was also notably reduced in eyes in POAG when compared with normotensive eyes (p = 0.001, fig 4).

Figure 4

NO2 levels in the aqueous humour of controls (259.20 (72.62) μmol/mg protein) and patients with glaucoma (156.82 (54.79) μmol/mg protein). Student’s t test for unpaired data: p = 0.001.

Association between cGMP in plasma and aqueous humour

A linear correlation between cGMP levels in plasma and aqueous humour (n = 43, r = 0.565, p<0.001) was found.

The association between plasmatic and aqueous humour cGMP concentrations was also statistically significant when calculated in the two groups separately (POAG group: n = 18, r = 0.514, p = 0.029; control group: n = 25, r = 0.558, p = 0.004) (fig 5).

Figure 5

Linear regression between cGMP plasma levels and cGMP concentrations in the aqueous humour.

The relation between plasma and aqueous cGMP significantly differed in glaucoma and control patients (overall test of coincidence, p = 0.003; comparison of slopes, p = 0.001).

Ocular perfusion pressure

Perfusion pressure was calculated in 23 of the 38 glaucoma patients and in 22 of the 46 controls. Mean ocular PP was similar in POAGs (51.21 (5.66) mm Hg) and in normals (53.26 (6.40) mm Hg; p = 0.262).

Considering glaucomas and controls together, there was a linear, positive correlation between cGMP plasma levels and PP (n = 45, r = 0.379, p = 0.01), as well as between plasmatic NO2 s and PP (n = 37, r = 0.339, p = 0.04). When the association of plasma cGMP and PP was explored in glaucoma patients and in controls, a positive correlation between the two variables was found in both groups (r = 0.249 and r = 0.423, respectively), but this relation did not attain statistical significance in POAG, whereas it was of borderline statistical significance in the control group (p = 0.050, fig 6). No statistically significant correlation was found between plasma NO2 levels and PP when glaucomas and controls were evaluated separately. No significant difference was found in the two groups of patients with respect to the association plasma cGMP with PP or plasma NO2 and PP.

Figure 6

Linear regession between cGMP plasma levels and ocular perfusion pressure (PP).


In the present study, decreased concentrations of NO2 and cGMP were found in the plasma and aqueous humour of glaucoma patients. Our data on PP show that plasma NO correlates with PP, in that higher levels of cGMP and NO2 are associated with higher PP values.

The lower plasma levels of NO indicators in patients with POAG could reflect an impaired balance of endothelium derived mediators in glaucoma, as suggested by other evidence.21,30

Decreased plasma cGMP seems to correlate with lower cGMP in the aqueous humour. As NO has a very short half life (2–30 s1), and production of NO in the anterior chamber has been documented,24,31,32 NO in aqueous humour should mainly derive from local formation. For this reason, the association found between plasma and aqueous humour cGMP is interesting, and suggests that a basic disorder in NO balance may act both systemically and locally. Our findings in the aqueous humour agree with those of other authors,25 and with the report of decreased NOS reactivity in the aqueous outflow pathway of postmortem glaucomatous eyes.33 Such alterations, together with pharmacological studies showing that NO donors reduce IOP elevating NO2 in the anterior chamber,24 suggest that NO might be involved in the pathogenesis or regulation of IOP in POAG.

Higher IOP could alter NO concentration in the aqueous humour, but not in peripheral blood. Ocular antihypertensive medications such as β blockers could have an influence on NO levels in the aqueous humour and plasma. Although the β blocker propanolol inhibited the nitrite production evoked by a β adrenergic receptor agonist, it had no inhibitory effect either on NO formation induced by mediators other than β adrenergic agonists, or on basal production of NO.34 Moreover, NOS reactivity of glaucomatous eyes which had never received any antiglaucomatous medications showed severe alterations, similar to those found in the treated eyes.33 The effects of systemic β blockers on plasma NO levels are still unclear, even if some β blockers, such as nebivolol, are known to induce arterial and venous dilatation by greatly increasing cGMP.35,36

Impaired NO formation can have a dual negative effect in glaucoma patients by acting on IOP and ocular blood supply, as shown by the relation between plasma cGMP or NO2 and PP. The correlation between the two parameters found in controls indicates that decreased NO levels are associated with lower PP. Nitric oxide could influence PP modulating both IOP and blood pressure. Lower PP is strongly associated with an increased prevalence of POAG.37 An impaired ONH autoregulation makes the glaucomatous optic nerve more sensitive to decreased PP.19 Recently, NO has been involved in choroidal autoregulation.38 In our study, the relation between plasma cGMP or NO2 and PP did not attain statistical significance in glaucomas. This could be because of the small data set, or to glaucoma therapy, that makes IOP less affected by physiological stimuli. Otherwise, an altered NO formation could modify NO influence on PP. This hypothesis would fit well with the report that glaucoma patients receiving nitrate therapy for systemic conditions have less progression of glaucomatous optic neuropathy and visual field loss than patients who do not take these compounds.39


Research winner, the Italian selection of the Merck Sharp & Dohme International Award 2002.


View Abstract