Statistics from Altmetric.com
Editor,—Subhyaloid or subinternal limiting membrane haemorrhage (sub-ILM) is usually caused by diabetic retinopathy,1 hypertensive retinopathy,2retinal artery macroaneurysm,2 Valsalva retinopathy,3 Terson's syndrome,2 or blood dyscrasias.4 Bacterial meningitis has been associated with subhyaloid or sub-ILM haemorrhage.5 6 We report a case of subhyaloid or sub-ILM haemorrhage in a young boy with meningococcal meningitis, and discuss the possible mechanisms and the management of this condition.
A 13 year old boy was admitted to the paediatric ward with a 24 hour history of progressive drowsiness, headache, neck stiffness, nausea, and vomiting. On examination, he was confused, agitated, and pyrexial. There were petechial skin haemorrhages on both arms and legs. The blood pressure was normal and all peripheral pulses were present. There were no focal neurological signs. Funduscopy showed no obvious abnormalities. Lumbar puncture was contraindicated because of the risk of raised intracranial pressure. A clinical diagnosis of meningococcal meningitis was made. He was treated with intravenous benzylpenicillin and cefotaxime, and he was sedated and ventilated for 48 hours.
Gram stain and microscopic examination of smears from petechial skin lesions revealed intracellular Gram negative diplococci consistent withNeisseria meningococcus infection. Blood culture and polymerase chain reaction studies were negative. The prothrombin time was prolonged on the first two consecutive days to 19 seconds (normal 11.5–15 seconds). The activated partial thromboplastin time and the platelet count were normal. His condition improved rapidly. He was discharged home on day 6, and he returned to hospital for intravenous ceftriaxone therapy daily for another 4 days.
On day 9, he noticed poor vision in the right eye on inadvertently covering his left eye. Ophthalmological examination revealed visual acuity to be counting fingers right eye and 6/5 left eye. Anterior segments were normal. Funduscopy of the right eye revealed a round, well circumscribed, dome-shaped haemorrhage overlying the posterior pole consistent with a subhyaloid or sub-ILM haemorrhage (Fig 1). Conservative management was decided upon because of his young age, recent severe illness, and good vision in the left eye. By the fourth week, there was contraction in size of the haemorrhage. By the 16th week, the right preretinal haemorrhage had almost completely cleared and vision had improved to 6/5. Fundus fluorescein angiography revealed no retinal or choroidal abnormality (Fig 2).
As far as we know, there have been two previously reported cases of subhyaloid or sub-ILM haemorrhage associated with bacterial meningitis.5 6 The exact mechanism of this haemorrhagic response is not fully understood. Valsalva retinopathy has been suggested.5 Coughing and vomiting cause a rapid rise in intravenous pressure within the eyes. This can cause spontaneous rupture of superficial retinal capillaries.3
Elevation of intracranial pressure is a potential complication of meningococcal meningitis.7 A sudden increase in intracranial pressure can rupture the epipapillary and peripapillary capillaries, known as Terson's syndrome.2 In this case, lumbar puncture was clinically contraindicated, but raised intracranial pressure may have occurred.
Clotting disorders have been reported to cause sub-ILM haemorrhage.8 The mild prolongation of the prothrombin time in this patient could be due to the consumption of coagulation factors as a result of the release of bacterial endotoxin.7 This mild clotting defect may have caused the preretinal haemorrhage in this case.
Spontaneous resorption of subhyaloid or sub-ILM haemorrhage caused by Valsalva retinopathy usually occurs without sequel.3 Pars plana vitrectomy has been recommended for dense preretinal haemorrhage resulting from diabetic retinopathy, as spontaneous resorption is usually prolonged in these patients.1 Laser puncturing of the posterior hyaloid face or the internal limiting membrane has been described as an alternative to vitrectomy.9 Ulbiget al 10 recently reported the results of a series of premacular subhyaloid haemorrhages treated with pulsed Nd:YAG laser. Nevertheless, this procedure is not without complications. They reported one case of macular hole formation and one case of retinal detachment after the procedure.
We feel that premacular subhyaloid or sub-ILM haemorrhage caused by factors other than diabetic retinopathy may be managed conservatively in the first few months. Our case demonstrates that a good visual outcome can be achieved with conservative management.