Intended for healthcare professionals


Thrombolysis in patients with diabetes

BMJ 1995; 310 doi: (Published 07 January 1995) Cite this as: BMJ 1995;310:3
  1. Helen Ward,
  2. John S Yudkin
  1. Senior house officer in medicine Professor of medicine Department of Medicine, Whittington Hospital, London N19 5HT

    Withholding treatment is probably mistaken: patients should be given a choice

    Any junior doctor treating a patient with diabetes mellitus and an acute myocardial infarction faces a dilemma. Lists of cautions and contraindications for thrombolytic treatment usually include diabetic retinopathy. The reasonable fear of precipitating a vitreous or retinal haemorrhage helps to explain why fewer diabetic than non-diabetic patients are given thrombolysis.1 2 Funduscopy is not, however, easy in a brightly lit receiving room after the administration of opiates. Even after mydriatic drops are given it may not be possible definitely to exclude changes in the eye. The next hurdle to face after making the decision to give thrombolysis—or not—is to justify one's actions on the post-take ward round.

    The British National Formulary states that diabetic retinopathy is a contraindication to thrombolysis, although this will be changed to a caution in future editions. The datasheets from drug manufacturers vary from making no mention of diabetes (anistreplase, Boehringer) through advising special caution in the presence of diabetic proliferative retinopathy (alteplase, Boehringer) to stating that thrombolysis is contraindicated in severe diabetes mellitus (streptokinase, Hoechst) or in diabetic retinopathy (streptokinase, Pharmacia). Junior doctors must find it difficult to give a drug when its use is directly contraindicated in the British National Formulary.

    Against that background the lack of published case reports is surprising. We have been able to find one account of bleeding from retinopathy in a single diabetic patient after thrombolysis3 and one other of ocular haemorrhage after streptokinase in a patient without diabetes.4 In neither case was there any long term effect on vision. The Committee on Safety of Medicines has received one report of subconjunctival haemorrhage associated with streptokinase. In a published overview of fibrinolytic trials in patients with myocardial infarction the proportionate reduction in 35 day mortality was slightly, but not significantly, greater in diabetic patients (136/1000 v 173/1000; 21.7%) than in non-diabetic patients (87/1000 v 102/1000; 14.3%).5 These figures imply that, for every 1000 diabetic patients treated, 37 patients survive who would otherwise have died. The overview of fibrinolysis found no evidence of excess bleeding or stroke in the diabetic patients. One small study suggested an excess risk of haemorrhagic complications in diabetic patients aged over 75,6 but in an analysis of over 9000 patients treated with thrombolysis, of whom a tenth had diabetes, complication rates were similar in the diabetic and non-diabetic patients.7

    Among the large trials of thrombolytic treatment only that conducted by the Gruppo Italiano per lo Studio della Strepto-chinasi nell'Infarto Miocardico included haemorrhagic diabetic retinopathy as a contraindication to treatment,8 while the second9 and third10 international studies of infarct survival and the study by the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico11 made no mention of diabetes, with or without retinopathy, in their exclusion criteria. In these trials alone morethan 80000 patients, of whom around 11% had diabetes, received thrombolytic treatment, without any reports of detrimental effects in their eyes. In a subgroup analysis of the thrombolysis and angioplasty in myocardial infarction trial no retinal haemorrhages were seen in 148 diabetic patients, 7% of whom had documented retinopathy.12

    Patients with diabetes have a poor outcome after myocardial infarction, yet—as for some other groups with a poor prognosis—they are less likely to receive thrombolytic treatment.2 We believe that in the light of the proved benefits of thrombolysis in patients with acute myocardial infarction the treatment should not be withheld on the basis of the existence of diabetic retinopathy.4 Any increase in risk seems to be small. Those who support the empowerment of patients might also suggest that this is a decision in which it is appropriate for the patient to have a voice.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract