An ophthalmologist referred a 5-year-old boy because he was concerned about physical abuse. During what he had anticipated would be a routine "red eye" evaluation, the ophthalmologist had observed subconjunctival and retinal hemorrhages. The only explanation the mother had offered was that a bag of potato chips had hit the child's eye.
Both mother and son seemed reserved on arrival. The boy reported that he had been hit in the left eye by a bag of potato chips thrown by his mother 2 days ago. A red eye had soon developed and became progressively worse. He had 1 day of fever and a slight cough. His mother also reported that he had been somewhat less active during the past few days and that perhaps he had been bruising more easily. He had not been taking any medications. His medical history was unremarkable and his immunizations were upto-date.
On physical examination, the boy was quiet and cooperative. His temperature was 37.5°C, his heart rate was 111 beats per minute, his respiratory rate was 24 breaths per minute, and his blood pressure was 102/53 mm Hg. In the left eye, approximately 50% of the sclera was obscured by hemorrhage. The presence of bilateral, flame-shaped retinal hemorrhages was confirmed and pale conjunctiva and nailbeds were noted. He also had gingival ecchymoses around a central incisor. Small, bilateral anterior cervical lymph nodes were palpable. His heart had a regular rate and rhythm with a II /VI systolic murmur at the left sternal border. The lungs were clear to auscultation. The abdomen was soft, nontender, and without palpable masses. The results of the neurologic examination were normal. A few bruises were noted on the pretibial area and he had several petechiae on his chest.
When the history is not consistent with the injury found, it is appropriate to be concerned about physical abuse. Retinal hemorrhages have been reported in 40% to 95% of children who have suffered shaken baby syndrome.1 Unfortunately, there are no pathognomonic findings to help distinguish retinal hemorrhages of abuse from those due to other etiologies. However, retinal hemorrhages of abuse are generally found in children younger than 3 years because their relatively large head, weak neck muscles, and immature brain myelination make them more susceptible to injuries from shaking.2
Retinal hemorrhages that are more severe (associated with optic atrophy), retinal folds, and retinal detachments are more indicative of abuse.23 Although often bilateral, unilateral retinal hemorrhages from abuse have been reported.43
The biomechanics of shaken baby syndrome and retinal hemorrhages are not completely understood. Previously, it was thought that the rapid acceleration and deceleration associated with violent shaking was sufficient to account for the findings of shaken baby syndrome.6 However, more recent studies suggest that a sudden deceleration caused by impact on a surface is required to reach the "injury threshold" seen in children with shaken baby syndrome, and that the entity should more correctly be called shaken impact syndrome.1,7
Several mechanisms have also been proposed to explain retinal hemorrhages. The forces of angular deceleration generated during violent shaking may be sufficient to cause retinal hemorrhages by causing vitreous traction on the retina.8 Increased thoracic or intracranial pressure leading to increased retinal venous pressure may also play an important role in the development of retinal hemorrhages.7'9
For our patient, however, his age and relatively well appearance argued against an abusive mechanism. To sort out the differential diagnosis, a battery of tests was administered, including a complete blood cell count with differential, a prothrombin time and partial thromboplastin time, and a blood type and antibody screen.
Although not relevant for our patient, birth is probably the most common cause of retinal hemorrhages. The timing of the ophthalmologic is important in detecting retinal In two large studies of term infants, one-third of the infants examined soon after birth had retinal hemorrhages. retinal hemorrhages may be seen after cesarean section, they are more commonly assowith vaginal delivery.10
In a study conducted by Emerson et al.,11 most hemorrhages (86%) resolved by 2 weeks birth, and only a single subretinal hemorpersisted at 4 weeks. The authors concludthat retinal hemorrhages in infants older than month are probably not related to birth. there have been several cases of neonaretinal hemorrhages that persisted for as long 6 months.12,13 So, although it might be difficult distinguish retinal hemorrhages secondary to birth from those associated with other factors in the early newborn period, we continued to consider other causes in our 5-year-old patient.
Although retinal hemorrhages may occur as a result of unintentional trauma, they are almost exclusively the result of significant mechanisms of injury, such as major head injuries sustained in motor vehicle collisions.1'2,7 In a recent case series of retinal hemorrhages from "household" trauma, the mechanisms of injury were not trivial.14 Two of the children hit a concrete basement floor: one had fallen down the stairs in a walker and the other had fallen from the top step over the side of the stairs. In addition, the retinal hemorrhages were unilateral and resolved quickly in all children.
Hematologic disorders, especially those associated with thrombocytopenia or clotting dysfunction, may cause retinal abnormalities. There are reports of retinal hemorrhages associated with leukemia,15 aplastic anemia,16 protein C deficiency,17 and sickle cell disease.18 We anxiously awaited our patient's blood test results to screen for the presence of a blood dyscrasia as the etiology of his retinal hemorrhages.
A host of infectious agents have been associated with retinal hemorrhages, including Rickettsia rickettsii, cytomegalovirus, falciparum malaria, and human immunodeficiency virus.2 Septic emboli secondary to rheumatic mitral and aortic valvulitis, subacute bacterial endocarditis, and syphilitic aortitis can also result in retinal abnormalities. Our patient had had a mild fever, but only a few upper respiratory tract symptoms and no systemic complaints. He had no history of a tick bite, and had not traveled recently. Although it is important to keep infectious etiologies in mind, it seemed unlikely that this was the cause of his retinal hemorrhages.
Chronically elevated or sudden extreme elevations in mean arterial blood pressure can cause flame or splinter hemorrhaging of the retinal vessels. A funduscopic examination is a key component of the evaluation of the patient with hypertension. In addition, the fundi should be carefully examined in patients presenting with symptoms suggesting hypertension, such as intermittent headaches or chest pain. Our patient had neither a history of hypertension nor symptoms that could be attributed to elevated blood pressure.
Although obviously not applicable in our patient, the role of cardiopulmonary resuscitation (CPR) in the development of retinal hemorrhages is controversial. There have been reports of children with retinal hemorrhages after CPR, but many may have been related to other causes or there was no funduscopic examination prior to the administration of CPR.9 Similarly, Gilliland and Luckenbach, in their postmortem ocular examination of children who had undergone CPR, found retinal hemorrhages only in those who had another known cause of retinal hemorrhages, such as trauma.19 In a prospective study by Odom et al., only 1 of the 45 patients had unilateral punctate retinal hemorrhages after CPR.20 It is unlikely that CPR alone is sufficient to cause retinal hemorrhages, and other etiologies should be investigated.
Retinal hemorrhages have been reported after general anesthesia, cardiac and intraocular surgery,2 juvenile glaucoma,21 and extracorporeal membrane oxygenation.22 Controversy exists as to whether seizure activity can predispose to retinal hemorrhage.23"25 Again, our patient had not experienced any of these predisposing conditions, so they were unlikely to play a causal role in mis situation.
The results of our patient's complete blood cell count were clearly abnormal: white blood cell count 17,000/ mL (2% bands, 4% neutrophils, 90% lymphocytes, 2% atypical lymphocytes, and 2% monocytes), hemoglobin 2.9 g/dL, hematocrit 8.6%, and platelets 7,000/ mL. The reticulocyte count was 1.1%. The electrolyte and liver function panels were normal, as were the prothrombin time and the partial thromboplastin time.
Arrangements were made for low-volume packed red blood cell and single-donor platelet transfusions, and the patient was admitted to the intensive care unit with a preliminary diagnosis of pancytopenia. The following day, a bone marrow aspirate and biopsy were performed. The biopsy specimen showed decreased cellularity with an absence of erythrocyte, granulocyte, and platelet precursors, thus suggesting the diagnosis of aplastic anemia.
The disorders of bone marrow failure include aplastic anemia, leukemia, and myeloproliferative disorders. In aplastic anemia, the bone marrow biopsy specimen typically shows hypocellularity and fatty infiltration of the marrow, distinguishing it from the other marrow failure syndromes. It is a rare disease in children, having an annual incidence of 2 per million.26 The causes are multiple. Aplastic anemia may be congenital (present at birth but not necessarily inherited), genetic (inherited but not necessarily present at birth), or acquired. The etiology often remains unknown despite an intensive search for an inciting cause. These cases are classified as "idiopathic" and constitute most of the acquired aplastic anemias.
Inherited aplastic anemia is often, but not universally, associated with phenotypic characteristics. These include Fanconi anemia, dyskeratosis congenita, Schwachman-Diamond syndrome, reticular dysgenesis, amegakaryocytic thrombocytopenia, familial aplastic anemias, preleukemia, myelodysplasia, monosomy 7, and nonhematologic syndromes such as Down syndrome. For the acquired causes, many associations have been made with various toxins (eg, benzene), drugs (eg, chloramphenicol, anti-inflammatory drugs, and antiepileptics), and viruses (eg, Epstein-Barr virus, hepatitis, parvovirus, and human immunodeficiency virus). In addition, aplastic anemia has been associated with thymoma, pregnancy, paroxysmal nocturnal hemoglobinuria, and preleukemia. Several reviews provide more detailed information.26,27
The symptoms of aplastic anemia depend on the severity of the pancytopenia. Because the life span of platelets is short, patients commonly present first with signs and symptoms of thrombocytopenia. These include petechiae, ecchymoses, epistaxis, and mucosal surface bleeding. Bacterial infections, oral ulcers, and fever may result from neutropenia. Anemia, characterized by pallor, fatigue, and tachycardia, is a late finding because the life span of the erythrocyte is longer than that of the other cell lines.
Peripheral blood cell counts nearly always show pancytopenia, and the reticulocyte count is decreased. Prothrombin time and partial thromboplastin time and antibody screens are generally normal. A bone marrow examination should be performed by both aspirate and biopsy because the aspirate alone can show false results or may be nondiagnostic. The biopsy generally shows hypocellularity, empty spicules, fat, reticulum cells, plasma cells, and mast cells. Results of peripheral blood and bone marrow chromosome tests are normal, in contrast to the chromosomal abnormalities seen in patients with Fanconi anemia and myelodysplasia.
Patients with aplastic anemia are treated initially with supportive care while the prospect of bone marrow transplant is assessed. Bleeding and infectious complications are treated with blood product support and the administration of broad-spectrum antibiotics. Blood products should be used sparingly to reduce antigenic exposure and should never be obtained from a family donor. Packed red blood cells should be washed or filtered prior to transfusion and it is preferable that platelets from a single donor be used.
The prognosis depends on the severity of the disease (categorized as "moderate," "severe," or "very severe" based on the blood cell counts at presentation) and the form of primary therapy. It is an oversimplification to state that the treatment options for primary therapy are bone marrow transplantation or immunotherapy (antilymphocyte globulin and antithymocyte globulin). The treatment decision is complex and based on a variety of case-specific features. There are some data, however, to suggest that pediatric patients who undergo transplant generally have better outcomes than do patients treated with immune therapy.26
Although it is true that one should consider child abuse when the history does not coincide with the results of the physical examination, in this case the history of "trauma" was a red herring. Retinal hemorrhages are not pathognomonic for child abuse, and other etiologies should be considered, especially in the child older than 3 years. Serious disease may present subtly, which emphasizes the need for a complete history and physical examination. Our patient's petechiae, gingival bleeding, pallor, and systolic murmur all pointed to a more systemic process.
1. Duhaime AC, Christian CW, Rorke LB, Zimmerman RA. Nonaccidental head injury in infants: the "shaken-baby syndrome." N Engl J Med. 1998;338:1822-1829.
2. Gayle MO, Kissoon N, Hered RW, Harwood-Nuss A. Retinal hemorrhage in the young child: a review of etiology, predisposed conditions, and clinical implications. J Emerg Med. 1995;13:233-239.
3. Levin AV. Ocular manifestations of child abuse. Ophthalmol Clin North Am. 1990;3:249-263.
4. Tyagi AK, Willshaw HE, Ainsworth JR. Unilateral retinal haemorrhages in non-accidental injury. Lancet. 1997;349: 1224.
5. Drack AV, Petronio J, Capone A. Unilateral retinal hemorrhages in documented cases of child abuse. Am J Ophthalmol. 1999;128:340-344.
6. Caffey J. On the theory and practice of shaking infants: its potential residual effects of permanent brain damage and mental retardation. Am J Dis Child. 1972;124:161-169.
7. Duhaime AC, Alario AJ, Lewander WJ, et al. Head injury in very young children: mechanisms, injury types, and ophthalmologic findings in 100 hospitalized patients younger than 2 years of age. Pediatrics. 1992;90:179-185.
8. Green MA, Lieberman G, Milroy CM, Parsons MA. Ocular and cerebral trauma in non-accidental injury in infancy: underlying mechanisms and implications for paediatric practice. Br J Ophthalmol. 1996;80:282-287.
9. The Ophthalmology Child Abuse Working Party. Child abuse and the eye. Eye. 1999;13:3-10.
10. Baum JD, Bulpitt CJ. Retinal and conjunctival haemorrhage in the newborn. Arch Dis Child. 1970;45:344-349.
11. Emerson MV, Pieramici DJ, Stoessel KM, Berreen JP, Galiano RF. Incidence and rate of disappearance of retinal hemorrhage in newborns. Ophthalmology. 2001;108:36-39.
12. Suzuki Y, Awaya S. Long-term observation of infants with macular hemorrhage in the neonatal period, jpn J Ophthalmol. 1998;42:124-128.
13. Beratis NG, Varvarigou A, Katsibris J, Gartaganis SP. Vascular retinal abnormalities in neonates of mothers who smoked during pregnancy. J Pediatr. 2000;136:760-766.
14. Christian CW, Taylor AA, Hertle RW, Duhaime AC. Retinal hemorrhages caused by accidental household trauma. J Pediatr. 1999;135:125-127.
15. Reddy SC, Menon BS. A prospective study of ocular manifestations in childhood acute leukaemia. Acta Ophthalmol Scand. 1998;76:700-703.
16. Mansour AM, Salti HI, Han DP, et al. Ocular findings in aplastic anemia. Ophthalmologica. 2000;214:399-402.
17. Hattenbach LO, Beeg T, Kreuz W, et al. Ophthalmic manifestation of congenital protein C deficiency. J Pediatr Ophthalmol Strabismus. 1999;3:188-190.
18. Gagliano DA, Goldberg MF. The evolution of salmonpatch hemorrhages in sickle cell retinopathy. Arch Ophthalmol. 1989;107:1814-1815.
19. Gilliland MG, Luckenbach MW. Are retinal hemorrhages found after resuscitation attempts? A study of the eyes of 169 children. Am ] Forensic Med Pathol. 1993;14:187-192.
20. Odom A, Christ E, Kerr N, et al. Prevalence of retinal hemorrhages in pediatric patients after in-hospital cardiopulmonary resuscitation: a prospective study. Pediatrics. 1997;99:e3. Available at www.pediatrics.org/ cgi/content/full/99/6/e3.
21. Jonas JB, Budde WM. Optic nerve head appearance in juvenile-onset chronic high-pressure glaucoma and normal-pressure glaucoma. Ophthalmology. 2000;107:704-711.
22. Varn MM, Donahue ML, Saunders RA, Baker JD, Smith CM, Wilson ME. Retinal examinations in infants after extracorporeal membrane oxygenation. J Pediatr Ophthalmol Strabismus. 1997;34:182-185.
23. Tyagi AK, Scotcher S, Kozeis N, Willshaw HE. Can convulsions alone cause retinal haemorrhages in infants? Br J Ophthalmol. 1998;82:659-660.
24. Sandramouli S, Robinson R, Tsaloumas M, Willshaw HE. Retinal haemorrhages and convulsions. Arch Dis Child. 1997;76:449-451.
25. O'Donohoe JM. Retinal haemorrhages and convulsions. Arch Dis Child. 1997;77:466.
26. Alter BP, Young NS. The bone marrow failure syndromes. In: Nathan DG, Orkin SH, eds. Nathan and Oski's Hematology of Infancy and Childhood, 5th ed. Philadelphia: W. B. Saunders; 1998:237-335.
27. Young NS, Maciejewski J. The pathophysiology of acquired aplastic anemia. N Engl J Med. 1997;336:1365-1372.