Educational paper : retinal haemorrhages in abusive head trauma in children

The Eyes Have It: How Critical are Ophthalmic Findings to the Diagnosis of Pediatric Abusive Head Trauma?

Pediatric abusive head trauma (AHT), still colloquially known as shaken baby syndrome, is a leading cause of morbidity and mortality among infants. Controversy has grown surrounding this diagnosis, and the specificity of the clinical findings-subdural hemorrhage, cerebral edema, and retinal hemorrhages-has been challenged. A literature search of peer reviewed publications on PubMed pertaining to the history, clinical, and pathologic features of AHT was conducted using the terms "shaken baby syndrome," "non-accidental trauma," "abusive head trauma," "inflicted traumatic brain injury," "shaken impact syndrome," and "whiplash shaken infant syndrome." Focus was placed on articles discussing ophthalmic findings in AHT. Retinal hemorrhages-particularly those that are too numerous to count, occurring in all layers of the retina (preretinal, intraretinal, subretinal), covering the peripheral pole and extending to the ora serrata, and accompanied by retinoschisis and other ocular/periocular hemorrhages-are highly suggestive of AHT, particularly in the absence of otherwise explained massive accidental trauma. Although the diagnosis has grown in controversy in recent years, AHT has well-documented clinical and pathologic findings across a large number of studies.

Patterns of retinal hemorrhage associated with pediatric cerebral sinovenous thrombosis

BACKGROUND

Cerebral sinovenous thrombosis (CSVT) has been proposed as an alternative cause of retinal hemorrhage (RH) in children being evaluated for abusive head trauma. This study investigated the prevalence and characteristics of RH in children with CSVT.

METHODS

The medical records of children >6 weeks of age with newly diagnosed CSVT and fundus examination by an ophthalmologist were examined retrospectively. Primary outcomes were presence and patterns of RH.

RESULTS

A total of 29 children (median age, 9 years; range, 7 weeks to 17 years) were studied. Of these, 5 (17%) had RH, in 4 of whom RH were peripapillary, superficial, intraretinal, and adjacent to a swollen optic disk. In the fifth child, who had meningitis, sepsis, and multiple cerebral infarcts, there were a moderate number of posterior pole intraretinal hemorrhages. Eighteen children (62%) had optic disk swelling. In 13 children, cerebrospinal fluid opening pressure was recorded (range, 27-59 cm H₂O). CSVT risk factors included meningitis, mastoiditis, and hypercoagulability.

CONCLUSIONS

RH in pediatric CSVT was uncommon. When RHs were present, the appearance matched RH patterns known to be caused by medical conditions, such as raised intracranial pressure and sepsis, also present in these children. These findings suggest that the RHs are due to these other causes and not directly to CSVT itself. In children with CSVT, if RHs are multilayered, extend beyond the peripapillary region into the rest of the posterior pole or retinal periphery, or occur in the absence of optic disk swelling, another etiology for the RH should be sought.

Yield and clinical efficacy of funduscopic examinations performed in the pediatric emergency room

Consultations by ophthalmologists to rule out papilledema are frequently requested by emergency room (ER) physicians. The clinical setting and optimal timing for examination are not well established, and the impact on patient management is unclear. We evaluated the yield of emergency funduscopic examinations, aiming at establishing the optimal timing and efficacy of the consultation. The medical records of all children aged 0-18 years referred for funduscopic examinations from the ER between June 2010 and May 2011 were reviewed. Of 19,772 ER visits, 1,920 (9.7 %) were seen by an ophthalmologist and 479 (2.4 %) to rule out papilledema. Headache (44.5 %) and head trauma (18.4 %) were the most common indications. Sixty percent of the 479 patients had been symptomatic for <24 h, all having normal eye exams. Only 6/479, with diverse etiologies, depicted papilledema. Among these six children, visual complaints associating with headache were statistically significant to suggest the presence of papilledema (p = 0.014). Seventy-one of the 479 children underwent neuroimaging studies despite normal funduscopic examinations. A single tumor case (medulloblastoma), symptomatic for weeks, had normal funduscopy.

CONCLUSION

The vast majority of ER consultations to rule out papilledema show normal findings, particularly among children with signs and symptoms lasting for <24 h. The yield of funduscopy may be higher if visual disturbances are reported. If neuroimaging studies are considered, emergency room ophthalmological consultation is probably not warranted, except for young infants with neurological signs and symptoms in whom retinal hemorrhages suggestive of physical abuse must be ruled out.