Aim: To determine the prevalence of retinal haemorrhages in infants presenting with convulsions and admitted to hospital, and to consider whether this finding indicates shaken baby syndrome.
Methods: Prospective study of children aged 15 days to 2 years admitted with a diagnosis of first convulsion over a 2-year period (May 2004–May 2006). All infants were examined by an experienced ophthalmologist using indirect ophthalmoscopy within 72 h of admission.
Results: 182 of 389 children seen in the accident and emergency department were admitted and two were found to have retinal haemorrhages. Both children were eventually diagnosed as being abused.
Conclusions: Convulsions alone are unlikely to cause retinal haemorrhages in children under 2 years of age.
Statistics from Altmetric.com
Convulsions are experienced by 40–70% of infants with shaken baby syndrome (SBS).1 SBS is an extremely serious form of abusive head trauma that occurs when a child is subjected to rapid acceleration, deceleration and rotational forces, with or without impact, resulting in a unique constellation of intracranial, intraocular and cervical spinal cord injuries.2 3 4 5 Although there is evidence that the described mechanism can produce neurotrauma in childhood, the exact aetiology and pathophysiology of the physical findings are the subject of debate in both the medical and the legal literature.6 7 8 9 10 11 12 13 14 Retinal haemorrhages are among the most common findings of SBS15 16 but may be associated with accidental head trauma as well as a variety of systemic illnesses.3 17 Theoretically, convulsions may cause retinal haemorrhages due to the sudden rise in retinal venous pressure, following an increase in central venous pressure secondary to a rise in intrathoracic pressure.17 Therefore, the prevalence of retinal haemorrhages after convulsions remains unclear. Studies have concluded that the chance of developing haemorrhages after convulsions is low (under 0.05)18 19 20 but do not offer enough evidence to determine the prevalence with certainty. The question therefore arises: how often might we diagnose retinal haemorrhages in infants with first convulsion episodes without SBS? Our objective was to determine the prevalence of retinal haemorrhages in infants presenting with convulsions and admitted to hospital and to consider whether this finding indicates SBS.
What is already known on this topic
Convulsions are the presenting sign in many cases of shaken baby syndrome.
Retinal haemorrhages are among the most common findings in inflicted head trauma.
The chance of developing haemorrhages after convulsions is low (under 0.05).
What this study adds
The prevalence of retinal haemorrhages after a first convulsion is less than 0.017.
If retinal haemorrhages are detected, the possibility of non-accidental injury must be investigated.
This was a prospective study of children aged 15 days to 2 years old admitted with a diagnosis of first convulsion over a 2-year period (May 2004–May 2006). Criteria for admission were first afebrile seizure, complex febrile seizure, and a simple febrile convulsion with no focus of infection. Children with a history of trauma, with findings on physical examination suggestive of child abuse (contusions, abrasions and skeletal fractures) which had been further investigated, and others with recognised causes of retinal haemorrhages were excluded.
All infants were examined by an experienced ophthalmologist using indirect ophthalmoscopy within 72 h of admission after informed consent was obtained from their parents and the supervising medical team. The pupils of the patients were dilated with short-acting mydriatic agents. A detailed history for each child was taken to categorise the type of seizure. A complete neurological examination was performed, including neuroimaging and electroencephalogram where necessary, and a funduscopic examination was carried out by an expert ophthalmologist using indirect ophthalmoscopy. For children with positive findings (retinal haemorrhages and/or a neuroimaging finding inconsistent with the story reported by the caregivers), an in-depth investigation was carried out in order to rule out maltreatment, following our hospital’s protocol. The protocol includes – in addition to funduscopic examination to exclude retinal haemorrhages –at least one head computed tomography (CAT) and/or magnetic resonance imaging (MRI) study to rule out subdural and/or subarachnoid haemorrhage, a skeletal survey and/or bone scintigraphy to rule out fractures, and a full blood count and coagulation studies to rule out blood dyscrasia. Those patients suspected of having been abused are normally evaluated by a multidisciplinary team composed of paediatricians, psychologists, neurosurgeons, ophthalmologists and social workers. Their conclusions are forwarded to the government’s management agency for children and adolescents (DGAIA).
Data analysis was conducted using SPSS 12.0 software. The study population was described using frequencies and percentages for categorical variables and means, and standard deviations or medians and ranges for continuous variables. As we assumed that post-convulsion retinal haemorrhage is rare, we used the binomial test to estimate the statistical probability of such an event. If none of the “n” patients showed retinal haemorrhages, statistical analysis was undertaken using Hanley’s rule of three; we can be 95% confident that the chance of this event is at most 3 in n (that is, 3/n).21 22 23
Overall, 182 of 389 children seen in the accident and emergency department were admitted; 100 (54.9%) were boys and 82 girls (45.1%). Their ages ranged from 15 days to 24 months with a median age of 10.4 months. Ninety six children (52.7%) had afebrile generalised seizures, 13 (7.1%) had afebrile partial seizures, 58 (31.9%) had complex febrile seizures and 15 (8.2%) had simple febrile seizures. Fifty four patients (29.5%) presented three or more episodes of seizure during the first 24 h of hospitalisation, and for 52 patients (28.6%) the seizure lasted more than 5 min (3.8% were admitted with status epilepticus). The causes of convulsions are given in table 1.
Two of the children were found to have retinal haemorrhages. In both cases, the clinical examination and investigations resulted in a diagnosis of abuse. Moreover, in case 1, the baby-sitter admitted to having shaken the infant before the seizure started (as reported in Barcenilla et al15). Table 2 provides details of these two patients. Figure 1 is a photograph of the funduscopic examination in case 2, showing a retinal haemorrhage in the posterior pole of the left eye above the macular region, a broader subhyaloid haemorrhage in resolution below the macular region and a small haemorrhage next to the papilla inferior border.
No retinal haemorrhages were found in any other children studied. Therefore, using Hanley’s rule of three, we can be confident to an upper limit of 95% that the chance of retinal haemorrhages occurring as a result of convulsions alone is at most 0.0167.
This study aimed to ascertain whether convulsions alone could produce retinal haemorrhages. In spite of the fact that 28.6% of cases had prolonged convulsions lasting more than 5 min (4% of which were status epilepticus) and 30% had repeated convulsive episodes, retinal haemorrhages were not detected in any of these children. Therefore, in our study, after excluding the two clear-cut cases of abuse, none of the 180 children under the age of 2 were found to have retinal haemorrhages within 72 h of their admission to hospital following convulsion. These results show that only rarely (prevalence lower than 0.017) is a first convulsion associated with retinal haemorrhages.
Previously, Sandramouli et al,18 studying 13 children under 2 years of age with convulsions, and Tyagi et al,19 studying 32 children, found no retinal haemorrhages, but Mei-Zahav et al20 found one case in a sample of 153. This child had a febrile seizure with no indication of abuse on investigation. Another study by Aghadoost et al24 describes two patients with retinal haemorrhages and convulsions but does not give their ages or the causes of the seizures. By combining the results of our study with those of previous studies (except for that by Aghadoost et al24 because of its limitations), we observe that there was only one case of retinal haemorrhage in the 378 non-abused reported patients. Therefore, the prevalence of retinal haemorrhages in the setting of first seizure was 0.0026 (95% CI 0.00007 to 0.015). This prevalence is much lower than that observed in abused children (0.76 in the King et al study).4 In consequence, the finding of retinal haemorrhage after a convulsive episode should trigger a careful search for other causes, especially child abuse.
In conclusion, convulsions alone are unlikely to cause retinal haemorrhages in children under 2 years of age. Therefore, if retinal haemorrhages are detected, investigation into the possibility of non-accidental injury is essential to safeguard the patient and his or her siblings. In addition, our study highlights the importance of undertaking a funduscopic examination of all infants with an apparently unprovoked seizure as retinal haemorrhages suggest abuse.
Competing interests None.
Provenance and Peer review not commissioned; externally peer reviewed.
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.