Optic Nerve Damage in Shaken Baby Syndrome Detection by b-Amyloid Precursor Protein Immunohistochemistry Aaron M. Gleckman, MD; Richard J. Evans, MD; Michael D. Bell, MD; Thomas W. Smith, MD

● Background.—Rapid acceleration-deceleration of an infant’s head during intentional shaking should in theory exert stretch or shear forces upon the optic nerves sufficient to cause axonal injury. b-Amyloid precursor protein (bAPP) immunohistochemistry recently has been shown to be a highly effective method for identifying diffuse axonal injury in the brains of infants with shaken baby syndrome. In this study, we investigated the utility of b-APP in identifying optic nerve damage in infants who have sustained fatal whiplash shaking. Materials and Methods.—b-Amyloid precursor protein immunohistochemistry was performed on formalin-fixed, paraffin-embedded sections of eyes (including optic disc and distal optic nerve) from infants less than 1 year of age with shaken baby syndrome (5 cases), combined shaken baby syndrome/blunt head trauma (3 cases), and ‘‘pure’’ blunt head trauma (1 case). Nontraumatic control cases included infants who died of suffocation (1 case), sudden infant death syndrome (1 case), and positional asphyxia (1

case) and an enucleation from a child with a retinoblastoma (1 case). Matched hematoxylin-eosin– and neurofilament-stained sections were used for comparison. Results.—Three of the 5 shaken baby cases and all 3 combined shaken baby/blunt head trauma cases had optic nerve axonal injury identified by the presence of strongly b-APP–immunoreactive beaded or swollen axonal segments. Axonal injury could not be detected in the corresponding hematoxylin-eosin– or neurofilament-stained sections. Optic nerve axonal injury was not seen in the case involving pure blunt head trauma or in the nontraumatic control cases. Conclusions.—Optic nerve axonal injury is a prominent feature of intentional fatal whiplash head trauma in infants less than 1 year of age. b-Amyloid protein precursor immunohistochemistry appears to be the most effective method for demonstrating axonal damage in the optic nerve. (Arch Pathol Lab Med. 2000;124:251–256)

S

elongation (stretch) of the optic nerve, resulting in morphologic abnormalities that are identical to those seen in posttraumatic diffuse axonal injury in the brain.3 To our knowledge, this type of optic nerve injury has not been described previously in infants with SBS. One likely explanation is that conventional hematoxylin-eosin or silver staining techniques may not be sensitive enough to detect more subtle degrees of axonal injury. b-Amyloid precursor protein (b-APP) immunohistochemistry recently has been shown to be a very sensitive and reliable method for detecting diffuse axonal injury in craniocerebral trauma injuries,4–9 including those affecting infants with SBS or blunt head trauma (BHT).10,11 This technique is based on the fact that functional or physical damage to the axonal cytoskeleton causes interruption of axoplasmic flow, which in turn leads to accumulation of transported substances such as b-APP within damaged axonal segments.12,13 Since b-APP is synthesized by retinal ganglion cells and transported by fast axoplasmic flow into axons in the retinal nerve fiber layer and optic nerve,14–17 shearing or stretch injury to the optic nerve should result in focal accumulation of b-APP analogous to that observed in the brain in diffuse axonal injury. The present study was designed to evaluate whether axonal injury can be detected by b-APP immunohistochemistry within the optic nerves of infants with SBS or BHT. MATERIALS AND METHODS Case Material

haken baby syndrome (SBS) is defined as the constellation of injuries incurred as the result of intentional whiplash movement of an infant’s head.1 There may or may not be associated impact injury. Major findings include subdural and subarachnoid hemorrhages and diffuse axonal injury in the brain and retinal hemorrhages in the eyes. The finding of severe bilateral, often multilayered retinal hemorrhages is particularly suggestive of SBS. However, because retinal hemorrhages may also be observed in cases of accidental trauma, papilledema, sepsis, severe hypertension, and even as an artifact of resuscitation,2 investigators need to identify additional histologic abnormalities in the eye that may be more specific for SBS. The rapid acceleration-deceleration of the head that causes retinal hemorrhages in SBS should in theory exert stretch or shearing forces within the optic nerves sufficient to cause axonal injury. Axonal injury has been produced experimentally in guinea pigs by rapid and controlled

From the Office of the Chief Medical Examiner of Massachusetts, Boston, Mass (Drs Gleckman and Evans); the Medical Examiner Department, Metropolitan Dade County, Miami, Fla (Dr Bell); and the Presented at the 51st Annual Scientific Meeting of the American Academy of Forensic Sciences in Orlando, Fla, February 15–20, 1999. Reprints: Thomas W. Smith, MD, Department of Pathology (Neuropathology), University of Massachusetts Medical School, 55 Lake Ave N, Worcester, MA 01655. Arch Pathol Lab Med—Vol 124, February 2000

Archival paraffin-embedded blocks of autopsy sections from both right and left eyes were obtained from the Office of the Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al 251

Case Material and Results of Optic Nerve b-Amyloid Precurser Protein (b-APP) Immunohistochemistry*

Case No.

Age, mo/ Sex

1 2

1/M 3/M

Shaken Shaken

SBS SBS

HOM HOM

10 15

30 0

Y Y

1 1

1 1

3

4/M

Shaken

SBS

HOM

20

51

Y

1

1

4

10/F

Shaken

SBS

HOM

5

53

Y

2

2

5

2/M

Shaken

SBS

HOM

5

72

Y

2

2

6

2/M

Shaken, dropped SBS, BHT

HOM

10

68

Y

1

1

7

3/M

SBS, BHT

HOM

15

6

Y

2

1

8

8/M

Shaken, slammed Shaken, slammed

SBS, BHT

HOM

10

90

Y

1

2

History

Cause of Death

Manner of Death

Estimated Interval from Injury to Hospital Presentation, Course, Retinal b-APP b-APP min h Hemorrhage (Right Eye) (Left Eye)

Autopsy Findings

Bilateral SDH, CE, DAI Bilateral SDH, SAH, CE, DAI, knee contusions Bilateral SDH, SAH, CE, DAI Bilateral SDH, SAH, CE, DAI, thrombosis of internal jugular vein Bilateral SDH, SAH, CE, spinal cord infarction, posterior neck hem, periadventitial vertebral artery hem, distal tibial fxs Bilateral SDH, SAH, CE, DAI, subgaleal hem, rib fxs, chest contusions SDH, SAH, CE, subgaleal hem Skull fx, subgaleal hem, epidural hem, SDH, SAH, DAI, old wrist fracture Skull fx, DAI, rib fxs

Slammed against BHT HOM 10 0 Y 2 2 floor 10 1/M Overlay death SUF ACC Unknown 1 N 2 2 None 11 4/M Found dead in SIDS NAT Unknown 4 N 2 2 None crib 12 4/F Found wedged POS ASP ACC Unknown 3 N 2 2 None between couch and wall 13 24/F N/A RTB N/A N/A N/A N/A 2 N/A N/A * N/A indicates not applicable; SBS, shaken baby syndrome; BHT, blunt head trauma; SUF, suffocation; SIDS, sudden infant death syndrome; POS ASP, positional asphixia; RTB, retinoblastoma; HOM, homicide; ACC, accident; NAT, natural; Y, yes; N, no; plus sign, positive reaction; minus sign, negative reaction; SDH, subdural hematoma; CE, cerebral edema; DAI, diffuse axonal injury; SAH, subarachnoid hemorrhage; hem, hemorrhage; and fx, fracture. 9

2/M

Chief Medical Examiner, Boston, Mass; the Metropolitan Dade County Medical Examiner Department, Miami, Fla; and the Palm Beach County Medical Examiner Department, Palm Beach, Fla. A paraffin-embedded block of an eye from a surgical enucleation procedure (for retinoblastoma) was obtained from the Massachusetts Eye and Ear Infirmary (Boston, Mass). All autopsy and surgical specimens of eyes had been fixed in 10% buffered formalin for at least 1 day prior to sectioning. All sections included the optic disc and distal optic nerve. The sections were stained with hematoxylin-eosin, b-APP, and neurofilament protein (NF). The cases included 5 infants with well-established SBS who had no findings of cutaneous scalp contusion, subcutaneous or subgaleal hemorrhage, epidural hemorrhage, or skull fracture; 3 infants with a combination of findings of both BHT and SBS, socalled shaking-impact syndrome2; 1 case each of ‘‘pure’’ BHT, suffocation, sudden infant death syndrome (SIDS), and positional asphyxia; and 1 enucleation from a child with a retinoblastoma (Table). The infants with SBS (cases 1–5) ranged in age from 1 to 10 months. All SBS cases had a clinical history of whiplash shaking of the head and had autopsy findings of subdural, subarachnoid, and bilateral retinal hemorrhages. The recorded time intervals between the shaking episode and death ranged from 0 to 72 hours. The infants who died of both BHT and SBS (cases 6–8) and from pure BHT (case 9) ranged in age from 2 to 8 months. 252 Arch Pathol Lab Med—Vol 124, February 2000

All had subgaleal hemorrhage, subdural and subarachnoid hemorrhages, skull fractures, cerebral lacerations or contusions, and bilateral retinal hemorrhages. The recorded time intervals between their trauma and death ranged from 0 to 90 hours. The infants who died of suffocation, SIDS, and positional asphyxia (cases 10–12) ranged in age from 1 to 4 months. The child with the retinoblastoma (case 13) was 24 months old. The tumor did not involve the optic disc or nerve.

Immunohistochemistry Sections from the paraffin blocks were cut at 4 mm, heated at 608C for 30 minutes, then deparaffinized and hydrated through a series of xylene and alcohol preparations. The sections were then subjected to antigen retrieval by microwave exposure in 0.01 mol/L citrate buffer (pH 6.0) for 5 minutes at 800 W. Following replenishment of this solution, the slides were heated in a microwave oven for an additional 5 minutes and then allowed to cool for 20 minutes. Immunohistochemical staining for b-APP and NF was performed on a TechMate 1000 (Ventana Medical Systems, Tucson, Ariz) automated immunostainer using the avidin-biotin complex procedure. The b-APP monoclonal antibody (clone 22C11, Boehringer-Mannheim, Indianapolis, Ind) recognizes an N-terminal epitope (amino acids 60–100) of b-APP; it was used at a dilution of 1:160. The NF monoclonal antibody (clone 2F11, Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al

Dako Corporation, Carpinteria, Calif) recognizes the phosphorylated form of the 200-kd and 70-kd neurofilament subunits; it was used at a dilution of 1:200. Following a hydrogen peroxide block of endogenous peroxide and a serum blocking step, the slides were incubated with the primary antibody for 45 minutes, followed by brief buffer washes and incubation in a cocktail of biotinylated anti-mouse IgG/IgM and anti-rabbit IgG (Ventana) for 30 minutes. The sections were then washed, incubated in avidin-biotin complex (Ventana) for 30 minutes, washed, then reacted with diaminobenzidine and hydrogen peroxide to visualize the end product. Sections were counterstained with hematoxylin. Appropriate positive and negative controls were used for each antibody.

RESULTS Shaken Baby Syndrome Since diffuse low-level b-APP immunoreactivity was found to be normally present in the retinal ganglion cells, retinal nerve fiber layer, and distal optic nerve,14–17 we specifically defined axonal injury as the presence of focal highintensity b-APP axonal staining in association with the morphologic features of beading, fragmentation, or focal swellings. Using these criteria, we detected evidence of axonal injury within the optic nerves of 3 of 5 cases of SBS (cases 1–3, Table). In most cases, the damaged b-APP– positive axons were concentrated within a relatively narrow band in the distal optic nerve just caudal to the optic nerve head and lamina cribrosa (Figure 1, A and B). Scattered b-APP–positive axonal segments were also seen in more proximal regions of the optic nerve. Axonal injury could not be detected reliably in either the corresponding hematoxylin-eosin–stained section (Figure 1, C) or the section immunostained for NF (Figure 1, D). Definitive identification of axonal beading or small axonal swellings was largely masked by the strong diffuse NF immunoreactivity, which nonselectively labeled both injured and normal axons. Axonal injury was not associated with the presence of either hemorrhage or focal ischemic damage within the optic nerve parenchyma. Focal aggregates of strongly bAPP–positive axonal swellings were also present within the retinal nerve fiber layer in all 5 SBS cases (Figure 2) and were almost always adjacent to areas of retinal hemorrhage (cases 1–5, Table). Shaking-Impact Syndrome (BHT and SBS) b-Amyloid precursor protein immunostaining showed optic nerve axonal damage in at least 1 eye in all 3 cases of combined BHT/SBS (shaking-impact syndrome, cases 6–8; Table). The morphologic appearance and location of the injured b-APP–positive axons within the optic nerve were identical to those described in the SBS cases. These cases also exhibited focal b-APP–positive axonal swellings within the retinal nerve fiber layer associated with retinal hemorrhages. Blunt Head Trauma Optic nerve axonal injury was not observed in the case of BHT that was not associated with whiplash-type shaking (case 9). Nontraumatic Controls The eyes from infants who died of suffocation (case 10), SIDS (case 11), and positional asphyxia (case 12), as well as the enucleation for retinoblastoma (case 13), showed diffuse weak b-APP immunostaining in the retinal ganglion cell perikarya, nerve fiber layer of the retina, and Arch Pathol Lab Med—Vol 124, February 2000

distal optic nerve (Figure 3). None of the cases had strongly b-APP–immunoreactive beaded or swollen axonal segments indicative of axonal damage. COMMENT In this study, we have shown that axonal injury is present within the optic nerves of infants who died of SBS and shaking-impact syndrome. Furthermore, we have established that b-APP immunohistochemistry is the method of choice in demonstrating optic nerve axonal injury, inasmuch as both conventional hematoxylin-eosin staining and even NF immunohistochemistry were unable to produce convincing evidence of axonal damage. The fact that the majority of the infants who sustained whiplash shaking (SBS and shaking-impact syndrome) had optic nerve axonal injury supports our hypothesis that optic nerve axons are as susceptible as those in the brain to shear or stretch forces. Although optic nerve axonal injury was not seen in the one case of BHT unassociated with whiplash injury (case 9), the number of such cases in our series was clearly inadequate to draw any conclusion regarding the specificity of optic nerve axonal injury for SBS or shakingimpact trauma. Nevertheless, we believe that the presence of optic nerve axonal damage provides strong corroborative evidence that an infant has suffered a traumatic whiplash-type injury, especially in the presence of other autopsy findings such as retinal, subdural, or subarachnoid hemorrhages and diffuse axonal injury in the brain. We emphasize, however, that failure to observe axonal injury by b-APP immunohistochemistry does not necessarily exclude the presence of traumatic shear–type injury. Prior studies on cerebral axonal injury have shown that bAPP–immunopositive axons may not be observed if the survival period between injury and death is less than 2 hours.8 This finding may reflect both the relative insensitivity of the immunohistochemical technique in detecting very small amounts of b-APP and the fact that accumulation of b-APP is an evolving, time-dependent process.8,18 However, since the actual survival interval may not always be known, evaluation of the optic nerve using b-APP immunohistochemistry should not be omitted solely because the survival time is presumed to be too short. The detection of axonal injury by b-APP immunohistochemistry depends on a functional axoplasmic transport mechanism, which is an energy-dependent process.19 Either generalized hypoxia-ischemia due to cardiopulmonary arrest or diminished vascular perfusion of the optic nerve (via the ophthalmic artery) resulting from increased intracranial pressure could disrupt axonal energy production and impair axoplasmic flow within optic nerve axons. If such reactions occurred within 2 hours following trauma, an insufficient amount of b-APP would accumulate within an injured axon and would therefore be undetectable, even if survival extended well beyond the critical 2hour period.6 It is possible that this mechanism accounts for our failure to observe optic nerve axonal injury by bAPP immunohistochemistry in one of the SBS victims (case 5), who also sustained early generalized hypoxiaischemia. Axonal injury can be demonstrated by b-APP immunostaining immediately adjacent to areas of focal ischemia (infarction), where there is interruption of axoplasmic transport at the boundary between the normal and infarcted segments of the axon.20 Generalized hypoxia-ischemia, however, does not cause focal accumulation of b-APP Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al 253

Figure 1. Case 1. A, Axonal injury is indicated by the band of intense b-amyloid precursor protein (b-APP) immunoreactivity in the distal optic nerve just caudal to the lamina cribrosa (original magnification 310). B, At high magnification, b-APP–positive axonal beading and small swellings are observed (original magnification 3250). C, Axonal swellings are not seen in the corresponding hematoxylin-stained section of the same region as shown in B (original magnification 3150). D, Axonal injury is likewise difficult to detect in the corresponding neurofilament protein–stained section because of nonselective labeling of both normal and damaged axons (original magnification 3100). Figure 2. Case 3. b-Amyloid precursor protein immunostaining of retina shows extensive focal axonal injury within the nerve fiber layer adjacent to a recent hemorrhage (original magnification 350). Figure 3. Case 11. Distal optic nerve from an infant with sudden infant death syndrome demonstrates diffuse low-level b-amyloid precursor protein immunoreactivity within axons without evidence of beading or swelling. This appearance should not be interpreted as evidence of axonal injury (compare with Figure 1, B) (original magnification 3100).

254 Arch Pathol Lab Med—Vol 124, February 2000

Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al

within axons, because there is no physical disruption of the axonal cytoskeleton and any reduction in axoplasmic transport would more likely affect the axon in a diffuse rather than focal manner.10,21 Since infarcts were not present in any of the optic nerves in our cases, focal ischemia can be excluded effectively as a cause of the b-APP–positive axonal injury. Patients who have sustained severe head trauma, including infants with SBS and shaking-impact syndrome, may develop papilledema secondary to increased intracranial pressure. The question could be raised as to whether papilledema can cause axonal changes that could confound the identification of axonal injury due to shear or stretch forces. Experimental and human studies have shown that the histologic basis of papilledema is the combination of axonal swelling and interstitial edema in the optic nerve head.22–24 Some experimental models have demonstrated marked impairment of both fast and slow axoplasmic transport in papilledema.25,26 However, another study27 suggested that optic disc swelling is more likely due to a decrease rather than complete blockade of slow axoplasmic flow, with fast axoplasmic transport being relatively mildly affected. These data suggest that papilledema, at least in its early stages, is more likely associated with a functional, probably reversible abnormality in axoplasmic transport rather than with actual physical disruption of the axonal cytoskeleton leading to the formation of beading, fragmentation, or retraction balls. This would also be consistent with the clinical observation that most patients with papilledema do not experience permanent visual deficits, which might be expected if permanent axonal injury had occurred. We also emphasize that axonal swelling in papilledema predominantly affects the optic disc,22–24 whereas traumatic axonal injury occurs mainly in the optic nerve caudal to the disc, as we have shown in this study. For these reasons we believe that papilledema would not likely cause the type of axonal damage that could be confused with mechanical stretch or shear injury to the optic nerve. The present study has largely focused on the presence of axonal damage within the optic nerve, which appears to be most closely correlated with shear or stretch injury. However, we also observed that all SBS and shaking-impact cases had focal b-APP–positive axonal swellings in the outer nerve fiber layer of the retina. Since all of these cases also had retinal hemorrhages, it is less certain that axonal damage in this location can be entirely attributed to mechanical forces exerted on the retina itself. In fact, most of the injured axons in the nerve fiber layer were located adjacent to or within areas of retinal hemorrhage and retinal detachment, suggesting a probable cause-effect relationship. Aggregates of swollen axons within the retinal nerve fiber layer (cotton wool spots), with or without associated hemorrhage, can be seen a variety of other focal ischemic or hemorrhagic disorders affecting the retina.28 Thus, the presence of axonal damage within the retina appears to be a much less specific marker for a whiplashshaking type of injury as compared to damage within the optic nerve itself. In conclusion, optic nerve axonal injury appears to be a frequent finding in infants who have sustained acceleration-deceleration whiplash-type trauma. We have shown that b-APP immunohistochemistry is clearly superior to conventional hematoxylin-eosin histology and even NF immunostaining for detecting axonal injury in the optic Arch Pathol Lab Med—Vol 124, February 2000

nerve. Potential interpretive pitfalls relating to the use of b-APP immunohistochemistry can be avoided by taking the following factors into consideration: (1) b-APP immunoreactivity of optic nerve axons should be identified as axonal injury only if staining intensity exceeds normal background activity (if present) and is accompanied by the morphologic features of axonal beading or focal swelling; (2) b-APP immunohistochemistry may not show optic nerve axonal injury if the interval of survival is less than 2 hours; (3) adequate optic nerve vascular perfusion must have occurred during the initial period of survival to ensure adequate functioning of the axoplasmic transport mechanism; and (4) histologic sampling of the eye must include the optic disc and distal optic nerve, since this region appears to be most susceptible to shear or stretch injury. The authors thank Holly Goolsby and William Zane, MD, for their help in this project. References 1. Caffrey 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. 2. 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. 3. Tomei G, Spagnoli D, Ducati A, et al. Morphology and neurophysiology of focal axonal injury experimentally induced in the guinea pig optic nerve. Acta Neuropathol. 1990;80:506–513. 4. Sherriff FE, Bridges LR, Gentleman S, Sivaloganathan S, Wilson S. Markers of axonal injury in postmortem human brain. Acta Neuropathol. 1994;88:433– 439. 5. Gentleman SM, Nash MJ, Sweetin CJ, Graham DI, Roberts GW. b-Amyloid precursor protein (b-APP) as a marker of axonal injury after head injury. Neurosci Lett. 1993;160:139–144. 6. Sherriff FE, Bridges LR, Sivaloganathan S. Early detection of axonal injury after human head trauma using immunocytochemistry for b-amyloid precursor protein. Acta Neuropathol. 1993;87:57–62. 7. Gentleman SM, Roberts GW, Gennarelli TA, et al. Axonal injury: a universal consequence of fatal closed head injury? Acta Neuropathol. 1995;89:537–543. 8. McKenzie KJ, McLellan DR, Gentleman SM, Maxwell WL, Gennarelli TA, Graham DI. Is b-APP a marker of axonal damage in short-surviving head injury? Acta Neuropathol. 1996;92:608–613. 9. Geddes JF, Vowles GH, Beer TW, Ellison DW. The diagnosis of diffuse axonal injury: implications for forensic practice. Neuropathol Appl Neurobiol. 1997;23: 339–347. 10. Gleckman AM, Bell MD, Evans RJ, Smith TW. Diffuse axonal injury in infants with nonaccidental craniocerebral trauma: enhanced detection by b-amyloid precursor protein immunohistochemical staining. Arch Pathol Lab Med. 1999;123:146–151. 11. Shannon P, Smith CR, Deck J, Ang LC, Ho M, Becker L. Axonal injury and the neuropathology of shaken baby syndrome. Acta Neuropathol. 1998;95:625– 631. 12. Koo EH, Sisodia SS, Archer DR, et al. Precursor of amyloid protein in Alzheimer disease undergoes fast anterograde axonal transport. Proc Natl Acad Sci U S A. 1990;87:1561–1565. 13. Maxwell WL, Povlishock JT, Graham DL. A mechanistic analysis of nondisruptive axonal injury: a review. J Neurotrauma. 1997;15:419–440. 14. Loffler KU, Edward DP, Tso MOM. Immunoreactivity against tau, amyloid precursor protein, and beta-amyloid in the human retina. Invest Ophthalmol Vis Sci. 1995;36:24–31. 15. Morin PJ, Abraham CR, Amaratunga A, et al. Amyloid precursor protein is synthesized by retinal ganglion cells, rapidly transported to the optic nerve plasma membrane and nerve terminals, and metabolized. J Neurochem. 1993;61: 464–473. 16. Amaratunga A, Fine RE. Generation of amyloidogenic C-terminal fragments during rapid axonal transport in vivo of b-amyloid precursor protein in the optic nerve. J Biol Chem. 1995;270:17268–17272. 17. Chen ST, Patel AJ, Garey LJ, Jen LS. Expression of b-amyloid precursor protein immunoreactivity in the retina of the rat during normal development and after neonatal optic tract lesion. Neuroreport. 1997;8:713–717. 18. Blumbergs PC, Jones NR, North JB. Diffuse axonal injury and head trauma. J Neurol Neurosurg Psychiatry. 1989;52:838–841. 19. Ferreira A, Caceras A, Kosik KS. Intraneuronal compartments of the amyloid precursor protein. J Neurosci. 1993;13:3112–3123. 20. Yam PS, Takasago T, Dewar D, Graham DI, Mcculloch J. Amyloid precursor protein accumulates in white matter at the margin of a focal ischaemic lesion. Brain Res. 1997;760:150–157. 21. Baiden-Amissah K, Joashi U, Blumberg R, Mehmet H, Edwards AD, Cox

Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al 255

PM. Expression of amyloid precursor protein (b-APP) in the neonatal brain following hypoxic ischaemic injury. Neuropathol Appl Neurobiol. 1998;24:346– 352. 22. Tso MO, Fine BS. Electron microscopic study of human papilledema. Am J Ophthalmol. 1976;82:424–434. 23. Tso MO, Hayreh SS. Optic disc edema in raised intracranial pressure, III: a pathologic study of experimental papilledema. Arch Ophthalmol. 1977;95: 1448–1457. 24. Hayreh SS. Optic disc edema in raised intracranial pressure, V: pathogenesis. Arch Ophthalmol. 1977;95:1553–1565.

256 Arch Pathol Lab Med—Vol 124, February 2000

25. Tso MO, Hayreh SS. Optic disc edema in raised intracranial pressure, IV: axoplasmic transport in experimental papilledema. Arch Ophthalmol. 1977;95: 1458–1462. 26. McLeod D, Marshall J, Kohner EM. Role of axoplasmic transport in the pathophysiology of ischemic disc swelling. Br J Ophthalmol. 1980;64:247–261. 27. Radius RL, Anderson DR. Fast axonal transport in early experimental disc edema. Invest Ophthalmol Vis Sci. 1980;19:158–168. 28. Yanoff M, Fine BS. Ocular Pathology: A Text and Atlas. 3rd ed. Philadelphia, Pa: JB Lippincott Co; 1989.

Optic Nerve Damage in Shaken Baby Syndrome—Gleckman et al