Many causes of spontaneous hyphema, ranging from congenital to neoplastic, have been presented in the literature. To date, spontaneous hyphema has not been reported in the neonatal population, and hyphema, arising from a persistent pupillary membrane, has been discussed in only a handful of adult cases.1 We present a unique case of spontaneous hyphema in a neonate and discuss the presumed etiology, differential diagnosis, and management. To our knowledge, this case represents the youngest patient ever reported to have a spontaneous hyphema.
The patient was a girl born by cesarean section at 30 weeks' gestation with a birth weight of 1200 g. The mother had entered into spontaneous preterm labor secondary to intrauterine growth retardation and hydrops fetalis. The delivery was complicated by the neonate's respiratory arrest, requiring advanced cardiac life support for approximately 8 minutes. The first few days of life in the neonatal intensive care unit (NICU) were complicated by disseminated intravascular coagulation (DIC) and a grade III intraventricular hemorrhage requiring several transfusions of packed red blood cells and fresh frozen plasma. The DIC was diagnosed on the second day of life with the following laboratory results: PT 44.9 sec, PTT 70.8 sec, fibrinogen <50 mg/dL, and platelets 13 000 at birth. No protein C was drawn.
Fig 1: Low magnification external photograph showing hyphema OD.
Fig 2: Side view OD showing marked proptosis.
Fig 3: Prominent pupillary membrane OS
Fig 4: Diagram showing the vascular network.
On day 7, the NICU staff noted blood in the child's right eye. At that time, the DIC had improved (PT 15.8 sec, PTT 44.6 sec, fibrinogen 177 mg/dL, and platelets 71000), and workups for sepsis and the etiology of the DIC were underway. The girl was on a ventilator and received no antibiotics. Our initial examination of the right eye was significant for the marked appearance of proptosis (Figs 1, 2) with a complete hyphema obscuring the iris. The cornea was thinned centrally and appeared to be at significant risk for perforation. Intraocular pressure (IOP) by TonoPen (Mentor) was 39 mm Hg.
Examination of the left eye was significant for a prominent tunica vasculosa lentis with an engorged pupillary membrane across the iris plane and pupil (Figs 3, 4). Fundus examination of either eye was impossible (except for a positive red reflex OS) because of poor visibility through the anterior chamber. A B-scan ultrasound of the right eye showed no mass in the posterior segment. No vitreous hemorrhage or retinal detachment were appreciated. Ultrasound of the left eye was not performed.
Photos were taken and the patient was started on the following in the right eye: dorzolamide and apraclonidine, three times daily, (to lower intraocular pressure), prednisolone 1%, four times daily (for the hyphema), and erythromycin ointment, four times daily (to lubricate the cornea). A scleral shield was taped over the right orbit to help prevent globe rupture. Examination the next day revealed a reduced IOP to 33 mm Hg OD but no changes otherwise.
On day 11, the IOP was 18 mm Hg OD. The examination noted a partially consolidated clot in the anterior chamber of the right eye that appeared to be adherent to the corneal endothelium. The iris was partially visible, allowing visualization of a prominent tunica vasculosa lentis and engorged pupillary membrane. The apparent proptosis and corneal thinning appeared to be worse. A red reflex was present. The left eye appeared unchanged.
On day 12, the infant's Toxoplasma titers were positive and her physical status deteriorated. Total organ failure with a poor prognosis occurred. The ward team and the patient's parents decided to terminate life support, and the infant died. The family refused permission for autopsy.
Persistent pupillary membranes are the most common developmental abnormality of the iris and are present in about 95% of newborns and 20% of adults.2 The development of the tunica vasculosa lentis begins at the fifth week of gestation and is coincident with development of the anterior lens capsule. This vascular structure arises from capillaries originating in the hyaloid artery that sweep around the lens and anastomose with choroidal vessels, giving off branches that join the long ciliary arteries to form the pupillary membrane.2 The anterior portion of the tunica vasculosa lentas is continuous with the pupillary membrane. The pupillary membrane is fully developed by 9 weeks' gestation.3 During the fifth and sixth months of gestation, the pupillary membrane undergoes a remodeling, with regression of the central portion and the incorporation of the peripheral component into iris stroma. This regression of the vascular capsule normally is complete at the time of birth. Failure of the remodeling process occurring in the primary pupillary membrane may result in persistence of remnants, varying from a vascularized complete membrane to un vascularized strands stretching from collarette to collarette.1
Laidlaw and Bloom1 cite four known patients with spontaneous hyphema secondary to persistent pupillary membranes, the youngest of whom was 43 years old. To our knowledge, spontaneous hyphema has never been reported in the neonatal population; we think that this case presentation represents the youngest patient ever reported to have a spontaneous hyphema. Based on examination of the fellow eye, we feel the hyphema was secondary to a prominent pupillary membrane. We cannot rule out a persistent underlying coagulopathy as the etiology, despite normalizing clotting functions at the time of initial discovery of the hyphema and examination by the pediatric ophthalmology team. The most classic ocular finding in DIC is a serous retinal detachment.4 Other findings in DIC include retinal and vitreous hemorrhage, none of which was seen in this case.
Hyphemas in children most often result from trauma5 A forceps-induced hyphema is possible in the setting of a difficult delivery, but not likely in this case as the infant was delivered via Cesarean section. Spontaneous hyphema may occur in retinoblastoma,6 rctinoschisis, retrolental fibroplasia, persistent primary hyperplastic vitreous, blood dyscrasias, and other less-common conditions.7 For atypical presentations, workup should include complete blood count, coagulation screen, and hemoglobin electrophoresis.8 Spontaneous hyphema in infancy is recognized as a classic sign of juvenile xanthogranuloma.9 Zimmerman reported on a series of 28 cases of juvenile xanthogranuloma in which 85% who had ocular involvement were infants younger than 1; 64% were younger than 7 months.10
Our patient was managed with topical steroids and antiglaucoma drops. The IOP responded well to therapy, but the cornea continued to thin. We think the corneal compromise was secondary to the hyphema causing elevated IOP, resulting in a stretching and thinning of the cornea. With the increased proptosis came worsening exposure and dryness that may have exacerbated the thinning. The patient was not deemed a viable surgical candidate at the time because of her tenuous medical status. Had she lived, the visual prognosis would undoubtedly have been poor, due to the gross deformity of the cornea.
Hyphemas in the neonatal population place the infant at significant risk for amblyopia, both from the frank blood in the anterior chamber and from the residual corneal staining with hemoglobin and blood degradation products. The neonatal cornea may have a higher permeability for blood and driving blood through the endothelium and into the corneal stroma may be easier. Perhaps our threshold for treating hyphemas in neonates should be lower than for children and adults,11 both in lowering the IOP and in early anterior chamber washout, to avoid both amblyopia and corneal compromise.
1. Laidlaw DAH, Bloom PA. Spontaneous hyphema arising from a strand of persistent pupillary membrane. Eur J Ophthalmol. 1993;3:98-100.
2. Worgul BV. Lens. In: Tasman "W, Jaeger EA (eds). Duane's Foundations of Clinical Ophthalmology. Philadelphia, Pa: J.B. Lippincott; 1992;1:3,24.
3. Cook CS, Organice J, Jacobiec FA. Prenatal development of the eye and its adnexa. In: Tasman W, Jaeger EA (eds). Duane's Foundations of Clinical Ophthalmology. Philadelphia, Pa: JB. Lippincott; 1992;1:41.
4. Williams GA. Ocular manifestations of hematologic diseases. In: Tasman W, Jaeger EA (eds). Duane's Clinical Ophthalmology. Philadelphia, Pa: J.B. Lippincott; 1992;5:2,3.
5. Agapitos PA, Noel LP, Clarke WN. Traumatic hyphema in children. Ophthalmology. 1987;94:1238-1241.
6. Schuster SAD, Ferguson ED. Unusual presentations of retinoblastoma. South Med J. 1970;63:4-8.
7. Howard GM. Spontaneous hyphema in infancy and childhood. Arch Ophthalmol. 1962;68:615-620.
8. Murdoch IA, Dos Anjos R, Parsons JM, Calver DM. Spontaneous hyphema in childhood. Eur J Pediatr. 1991;150:717-718.
9. Harley RD, Romayananda N, Chan GH. Juvenile xanthogranuloma. J Pediatr Ophthalmol Strabismus. 1982;19:33-39.
10. Zimmerman LE. Ocular lesions of juvenile xanthogranuloma. Trans Am Acad Ophthalmol Otolaryngol. 1965;69:412.
11. Wilson FM. Traumatic hyphema: pathogenesis and management. Ophthalmology. 1980;87:910-919.