Letters

Omepraxole and ocular damage

BMJ 1998; 316 doi: http://dx.doi.org/10.1136/bmj.316.7124.67 (Published 03 January 1998) Cite this as: BMJ 1998;316:67

Facts of cases are unclear

  1. Paul Riordan-Eva, Consultant ophthalmic surgeona,
  2. Michael D Sanders, Consultant ophthalmologistb
  1. a West Kent Eye Centre, Farnborough Hospital, Orpington, Kent BR6 8ND
  2. b Department of Neuro-Ophthalmology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG
  3. c Department of Opthalmology, Harvard Medical School, 243 Charles Street, Boston, MA 02114, USA
  4. d University of California at Los Angeles, Los Angeles, CA 90073, USA
  5. e Zentralkrankenhaus Sankt-Jürgen-Strasse, 2800 Bremen 1, Germany

    Editor—In the report by Schönhöfer et al of visual loss related to the proton pump inhibitor omeprazole, several issues need clarification.1 Six of the nine cases are stated to have had “funduscopically confirmed” irreversible anterior ischaemic optic neuropathy. This clinical entity is not diagnosed solely on the basis of fundal findings but in the light of the nature and pattern of visual loss, the appearances of the optic disc in both eyes—an important clue being a relatively small optic disc in the unaffected eye2—and the results of fluorescein angiography. This is an important consideration because many forms of anterior optic neuropathy, including demyelinative or postviral optic neuritis and neoplastic optic nerve inflitration, can produce the same appearance of the optic nerve head as anterior ischaemic optic neuropathy. Indeed, the authors seem to be unclear whether demyelinative optic neuritis or anterior ischaemic optic neuropathy is associated with multiple sclerosis.

    It is also stated that two patients had no known risk factors for anterior ischaemic optic neuropathy. In fact, both had gastrointestinal ulcers, which have been shown in a prospective review of over 400 patients to be significantly associated with non-arteritic anterior ischaemic optic neuropathy.3 This in itself generates difficulty when attempting to associate a treatment for gastrointestinal ulcers with an increased risk of anterior ischaemic optic neuropathy.

    In their concluding sentence, the authors say that anterior ischaemic optic neuropathy is due to retinal artery ischaemia, whereas it is due to a perfusion deficit in the posterior ciliary circulation. There may well be an association between omeprazole and optic nerve dysfunction, but this paper fails to provide conclusive evidence of such an association or a clear understanding of its possible aetiology.

    References

    1. 1.
    2. 2.
    3. 3.

    Concerns on safety of drug are unwarranted

    1. Simmons Lessell, Professor of ophthalmologyc
    1. a West Kent Eye Centre, Farnborough Hospital, Orpington, Kent BR6 8ND
    2. b Department of Neuro-Ophthalmology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG
    3. c Department of Opthalmology, Harvard Medical School, 243 Charles Street, Boston, MA 02114, USA
    4. d University of California at Los Angeles, Los Angeles, CA 90073, USA
    5. e Zentralkrankenhaus Sankt-Jürgen-Strasse, 2800 Bremen 1, Germany

      Editor—Schönhöfer et al's article on ischaemic optic neuropathy from omeprazole1 may have created unwarranted concerns among doctors and patients about the safety of this drug.

      Several years ago, after allegations that visual impairment was a side effect of omeprazole, Astra invited me as a neuro-ophthalmologist to review case reports and decide if the evidence supported the allegations. Since then I have intermittently reviewed additional cases. The actual diagnoses in these cases were a hodgepodge of disorders (from uveitis to pre-syncope) but some represent instances of anterior ischemic optic neuropathy consequent to gastrointestinal bleeding, a complication of haemorrhage that was recognised many decades before omeprazole was introduced. I have yet to identify a single patient in whom drug toxicity caused visual dysfunction. The results of these case analyses were transmitted in detail at a meeting of the Committee for Proprietary Medicinal Products in Brussels in July 1994. The negative result of the “pharmacoepidemiological” study cited in the article also provides reassurance about the safety of omeprazole.

      Of the nine cases in the article, six had been discussed at the Brussels meeting. Case 3 was that of a patient terminally ill with metastatic cancer whose vague symptoms were never evaluated in life. Cases 1, 2, 4, and 5 were examples of a bilateral optic neuropathy in which a cause other than drug intoxication was likely. All of these patients received several drugs in addition to omeprazole. In case 6, the correct diagnosis was an inflammatory or demyelinative optic neuropathy.

      I suspect that case 7 is an example of anterior ischemic optic neuropathy and apparently omeprazole was the only drug that this patient received; however, as with the other cases it seems inappropriate to implicate omeprazole as the putative cause of the neuropathy. Idiopathic anterior ischemic optic neuropathy is by no means rare even in the absence of risk factors for vascular disease. It also is unlikely that a toxic optic neuropathy would affect only one eye.

      The hypothesis proposed by the authors for ocular damage induced by omeprazole seems ill founded. Omeprazole may well affect ATPases in non-gastric tissues, but there is no evidence that ATPases are inhibited in the optic nerve or retina or that such inhibition would have any role in the pathogenesis of anterior ischaemic optic neuropathy. The inhibitory effect of omeprazole on production of cerebrospinal fluid is irrelevant to optic nerve function. Drugs such as carbonic anhydrase inhibitors which reduce cerebrospinal fluid production do not impair optic nerve function and in fact are widely used in the long term to prevent optic nerve damage in glaucoma.

      References

      1. 1.

      Lack of causality holds true

      1. George Sachs, Professor of medicine and physiologyd
      1. a West Kent Eye Centre, Farnborough Hospital, Orpington, Kent BR6 8ND
      2. b Department of Neuro-Ophthalmology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG
      3. c Department of Opthalmology, Harvard Medical School, 243 Charles Street, Boston, MA 02114, USA
      4. d University of California at Los Angeles, Los Angeles, CA 90073, USA
      5. e Zentralkrankenhaus Sankt-Jürgen-Strasse, 2800 Bremen 1, Germany

        Editor—Schönhöfer et al discuss two new cases (cases 6 and 7) and attempt to link anterior ischaemic optic neuropathy with oral dosing of omeprazole.1 Two other patients (cases 8 and 9) had, they suggest, blurred vision as a result of oral omeprazole, but no data are presented for these. The other cases had been found by the experts of the Committee for Proprietary Medicinal Products to have ischaemia resulting from their primary disease, not from omeprazole. It is not possible to determine the cause of anterior ischaemic optic neuropathy in the new cases (6 and 7) from the data presented by Schönhöfer et al. That this neuropathy is unilateral shows that it is not related to a systemic drug, which would be expected to affect both eyes.

        The authors attempt to relate these cases to a role of the acid pump of the stomach in vascular smooth muscle. The gastric potassium-hydrogen ATPase is present in the stomach in large quantities and in the distal renal medulla at very low levels, but no other location for this pump has been found, even with modern molecular biological techniques. The reference quoted as providing evidence for the gastric potassium-hydrogen ATPase in vascular smooth muscle used various pharmacological agents before these more modern methods were available and drew incorrect conclusions. Schönhöfer et al say that inhibition of the pump supposedly present in the vasculature would result in intracellular acidification. The sodium-hydrogen exchanger would prevent such acidification even if the pump were present.

        The amino acid target, cysteine 813 of the catalytic subunit of the gastric potassium-hydrogen ATPase, is absent from other ATP driven ion pumps. The action of omeprazole depends both on accumulation due to a highly acidic space and on acid catalysed conversion to the active drug. These factors are absent in the vasculature. Hence, neither the target molecule nor the mechanism necessary for an action of omeprazole is present at the site suggested by Schönhöfer et al as explaining the anterior ischaemic optic neuropathy.

        The ocular events reported in cases 6 and 7 are probably not related to omeprazole. The absence of causality between the use of proton pump inhibitors and visual disturbances continues to hold true.

        References

        1. 1.

        Author's reply

        1. Peter S Schönhöfer, Professor of clinical pharmacologye
        1. a West Kent Eye Centre, Farnborough Hospital, Orpington, Kent BR6 8ND
        2. b Department of Neuro-Ophthalmology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG
        3. c Department of Opthalmology, Harvard Medical School, 243 Charles Street, Boston, MA 02114, USA
        4. d University of California at Los Angeles, Los Angeles, CA 90073, USA
        5. e Zentralkrankenhaus Sankt-Jürgen-Strasse, 2800 Bremen 1, Germany

          Editor—Riordan-Eva and Sanders miss a diagnostic differentiation between anterior ischaemic optic neuropathy and demyelinative optic neuritis. All patients had computed tomography or nuclear magnetic resonance imaging, or both, to exclude demyelinative or other diseases. Furthermore, linking anterior ischaemic optic neuropathy to gastrointestinal ulcers does not explain why the neuropathy was not seen in patients treated with other anti-ulcer drugs. Our hospital based monitoring system would certainly have signalled such cases, as it did with proton pump inhibitors.

          Lessell regards our hypothesis of omeprazole induced optic neuropathy as ill founded. However, he stated in a letter to Merck in 1994 that case 1 “might well be a toxic drug-effect of intravenous Antra.” It is surprising that the experts of the Committee for Proprietary Medicinal Products changed the diagnosis in case 6 from anterior ischaemic optic neuropathy to demyelinative optic neuropathy without contacting the hospital and checking the clinical records, which clearly argue against these experts' interpretation.

          Arteritis, also in nervous tissue, was described in toxicological studies with lansoprazole in dogs and discussed with respect to similar findings with other substituted benzimidazols.1 Various effects of omeprazole on human brain function are described.2 Vasoconstriction and disturbance in transcellular transport mechanisms may be involved due to inhibition of ATPases. Therefore, irreversible visual disorders (anterior ischaemic optic neuropathy, for example) may arise from ischaemia in the distribution territory of endarterioles and capillaries in the optic nerve tissue.

          Sachs reiterates what he has often said at Astra's press conferences. Drug companies and their experts use lack of causality in defending a product and to discredit reports of adverse effects. There is no causal explanation for the antidepressant effect of tricyclic agents, the antipsychotic effect of neuroleptics, or the sedative effect of barbiturates; there is only an association between the drugs and the desired result. This is sufficient to grant therapeutic efficacy. The same applies for unwanted effects. Visual disturbances and anterior ischaemic optic neuropathy occur in patients treated with proton pump inhibitors but not in those treated with other anti-ulcer drugs. This association does not disappear simply because experts declare that the events are not related causally. Such causality rating is misused as a bureaucratic safety measure.

          References

          1. 1.
          2. 2.
          View Abstract

          Sign in

          Log in through your institution

          Free trial

          Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
          Sign up for a free trial

          Subscribe