Article Text


Increased prevalence of ocular glaucomatous abnormalities in systemic sclerosis
  1. A Y K Chan1,
  2. D T L Liu2
  1. 1Department of Medicine and Therapeutics, Chinese University of Hong Kong, Hong Kong SAR
  1. Correspondence to:
    Dr D T L Liu
    Department of Ophthalmology and Visual Sciences, 7/F, Block B, Prince of Wales Hospital, Shatin, Hong Kong SAR 12345;
  1. Y Allanore3,
  2. A Kahan3,
  3. A Brezin4
  1. 2Department of Ophthalmology and Visual Sciences, Chinese University of Hong Kong, Hong Kong SAR
  2. 3Department of Rheumatology A, Paris V University, Cochin Hospital, AP-HP, 75014 Paris, France
  3. 4Department of Ophthalmology, Paris V University, Cochin Hospital, AP-HP, 75014 Paris, France

Statistics from

Dr Allanore and team members have conducted an inspirational study into the prevalence of ocular glaucomatous abnormalities in systemic sclerosis. However, it seems that they have overlooked a major methodological flaw about the definition of glaucomatous change pertaining to normal tension glaucoma (NTG). Unless further clarification can be offered, we cannot concur with the conclusion that glaucomatous neuropathy consistent with the vascular pathogenic hypothesis for NTG was dramatically more prevalent in patients with systemic sclerosis.1

Lee et al have revised the definitions of NTG of several major studies.2 Emphasis on maximal intraocular pressure (IOP) ⩽21 mm Hg and the importance of recognising the characteristic glaucomatous optic disc change or visual field defect have been implicated.2 Almost 89% of the studies/publications required the demonstration of a characteristic visual field defect on perimetry as a prerequisite for diagnosing NTG.2 In the present study, the above-mentioned key features have also been adopted as the defining criteria. Moreover, defining a case with a visual field mean deviation <−2 dB was arbitrary and differed significantly from our understanding of the neuropathological basis of visual field damage specific to NTG.

Araie and coworkers have indicated that NTG and the ordinary primary open angle glaucoma or high tension glaucoma (HTG) showed significantly different visual field damage.3 Visual field defects in NTG are more localised and predominant in the lower hemifield, whereas HTG has significantly more diffuse visual field damage.4–6 It has been demonstrated that mean deviation in perimetry is good measure for assessing the more diffuse visual field damage characteristic of HTG but not as good for pinpointing a localised defect such as that seen in NTG.7,8 Instead, pattern standard deviation or corrected pattern standard deviation were suggested as alternative indicators in representing the focal visual field defect in NTG.7,8 As a result, the authors’ conclusion about the relationship between NTG and systemic sclerosis may be based on an erroneous visual field index (mean deviation), which is neither sensitive nor specific for NTG.

Moreover, it should be pointed out that Allanore et al have adopted another arbitrary means of defining the IOP of the subjects recruited, which again showed marked disparity from our usual practice. The authors did not explain why phasing of the IOP was not undertaken given the fact that IOP shows diurnal variation, especially prominent in glaucomatous subjects such as those with NTG.9 Recording of only one IOP measurement may not be sufficient owing to the influence of this confounding factor.

Appropriate case definition lies at the heart of every epidemiological research on glaucoma and any deviation from the consensual definitions may inevitably skew or even imperil the validity of the data.9 In the interest of readers, we would be most grateful if the authors can provide us with more information about the rationale for the methodology used.


Authors’ reply

We thank Drs Chan and Liu for their comments about our article evaluating ocular glaucomatous changes in systemic sclerosis (SSc).

High intraocular pressure (>21 mm Hg) is undoubtedly known to be the main risk factor associated with glaucoma1; however, substantial evidence was provided recently to support a key role of vascular abnormalities in the pathogenesis of glaucoma. In particular, patients with normal tension glaucoma, who do not have the main risk factor of developing glaucoma (increased intraocular pressure), may also develop optic neuropathy, and numerous recent studies support the hypothesis that these lesions are associated with vasculopathies.2–4 These findings led us to investigate the prevalence of glaucomatous changes in SSc, a disease which is strikingly associated with generalised vascular involvement.

Although primary open angle glaucoma is well defined, normal tension glaucoma is more difficult to diagnose. Independently of intraocular pressure, glaucomatous changes are supported by optic disc cupping together with visual field defects.1 Thus, for the purpose of our comparisons between groups, we had to define cut off values for these two variables. For optic disc cupping, we chose a cut off point based on reported data5; we defined mild abnormalities as a c/d >0.3 and severe involvement as a c/d >0.7. For visual field, we also chose a mean difference <−2 dB according to reported data. Thus, the significant differences between SSc and matched controls for these measures allow us to suggest that patients with SSc have glaucomatous abnormalities as compared with our controls. Although there is no consensual definition of NTG, these results clearly suggest that patients with SSc have glaucomatous propensity. The continuing prospective standardised follow up of our patients and other series will quantify the precise risk factor of SSc for normal tension glaucoma.


View Abstract

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.