Guidelines for postmortem protocol for ocular investigation of sudden unexplained infant death and suspected physical child abuse.

Abstract

Postmortem examination is a cornerstone in identifying the cause of unexplained sudden death in children. Even in cases of suspected or known abuse, an autopsy may help characterize the nature of the abuse, which is particularly important in the forensic autopsy of children in the first 3 to 4 years of life when inflicted neurotrauma is most common. Forensic examinations are vital in cases that might otherwise be diagnosed as sudden infant death syndrome. The ocular autopsy in particular may demonstrate findings that were not appreciated on antemortem clinical examination. This protocol for postmortem examination of the eyes and orbits was developed to promote more consistent documentation of findings, improved clinical and forensic decision making, and more replicable and coherent research outcomes.

Full text

Page 1

ORIGINAL ARTICLE
Guidelines for Postmortem Protocol for Ocular
Investigation of Sudden Unexplained Infant Death and
Suspected Physical Child Abuse
M. G. F. Gilliland, MD,* Alex V. Levin, MD, MHSc, FRCSC,†‡ Robert W. Enzenauer, MD,§
Charles Smith, MD,¶ M. Andrew Parsons, MBChB, FRCPath,? Lucy B. Rorke-Adams, MD,**
James R. Lauridson, MD,†† G. Robert La Roche, MD,‡‡ Linda M. Christmann, MD,§§
Marcellina Mian, MD,‡ Jeffrey Jentzen, MD,¶¶ Kenneth B. Simons, MD,¶¶ Yair Morad, MD,??
Randell Alexander, MD, PhD,*** Carole Jenny, MD,††† and Tamara Wygnanski-Jaffe, MD‡‡‡
Abstract: Postmortem examination is a cornerstone in identifying
the cause of unexplained sudden death in children. Even in cases of
suspected or known abuse, an autopsy may help characterize the
nature of the abuse, which is particularly important in the forensic
autopsy of children in the first 3 to 4 years of life when inflicted
neurotrauma is most common. Forensic examinations are vital in
cases that might otherwise be diagnosed as sudden infant death
syndrome. The ocular autopsy in particular may demonstrate find-
ings that were not appreciated on antemortem clinical examination.
This protocol for postmortem examination of the eyes and orbits was
developed to promote more consistent documentation of findings,
improved clinical and forensic decision making, and more replicable
and coherent research outcomes.
Key Words: postmortem, autopsy, child abuse, shaken baby
syndrome, retinal hemorrhage, orbit
(Am J Forensic Med Pathol 2007;28: 323–329)
P
cases of suspected or known abuse, an autopsy may help
characterize the nature of the abuse, which is particularly
important in the forensic autopsy of children in the first 3 to
4 years of life when inflicted neurotrauma is most common.
Forensic examinations are vital in cases that might otherwise
be diagnosed as sudden infant death syndrome (SIDS).
The ocular autopsy may demonstrate findings that were
not appreciated on antemortem clinical examination for var-
ious reasons. Sometimes, the premortem examination is not
performed by an ophthalmologist; in other instances, al-
though the examiner was an ophthalmologist, funduscopy can
be limited because of the absence of pharmacologic dilation
of the pupils needed to visualize the peripheral retinal edges
(ora serrata). This circumstance usually occurs in comatose
children where the pupillary reaction is being used as indi-
cator of intracranial pressure. Other reasons for the premor-
tem failure to appreciate retinal pathology include opacities
of the ocular media, such as a vitreous hemorrhage, which
may obstruct the view, and the situation in which some
children die days, weeks, months, or years after the abusive
incident. In such cases, a timely ophthalmology examination
of the live child at the initial presentation may prove more
useful than later autopsy findings in which deterioration or
healing obscures the initial injury pathology. Yet, late au-
topsy according to this protocol may still be useful. In all fatal
cases, the autopsy findings should be correlated with the
ophthalmology examination in vivo.
In contrast to what may be commonly perceived, post-
mortem removal of the eyes and orbital contents is not
disfiguring as caps are placed under the lids and the orbit
filled with gauze or other material in preparation for viewing
ostmortem examination is a cornerstone in identifying the
cause of unexplained sudden death in children. Even in
Manuscript received January 27, 2006; accepted May17, 2006.
From the *Department of Pathology and Laboratory Medicine, The Brody
School of Medicine at East Carolina University, Greenville, North
Carolina; †Department of Ophthalmology and Vision Science, ‡Depart-
ment of Paediatrics, Suspected Child Abuse and Neglect (SCAN) Pro-
gram, ¶Department of Laboratory Medicine and Pathobiology, The
Hospital for Sick Children, University of Toronto, Toronto, Canada;
§Department of Ophthalmology, University of Tennessee Health Science
Center, Memphis, Tennessee; ?Ophthalmic Sciences Unit, University of
Sheffield, United Kingdom; **Department of Pathology, The Children’s
Hospital of Philadelphia, Philadelphia, Pennsylvania; ††Graphics, Mont-
gomery, Alabama; ‡‡Department of Ophthalmology, Dalhousie Univer-
sity and the IWK Health Centre, Halifax, Nova Scotia, Canada; §§Mon-
treal Children’s Hospital and McGill University Health Centre, Montreal,
Quebec, Canada; ¶¶Department of Pathology, Medical College of Wis-
consin, Milwaukee, Wisconsin; ??Assaf Harofeh Medical Center, Tel
Aviv University, Zrifin, Israel; ***University of Florida, Gainesville,
Florida; and Morehouse School of Medicine, Atlanta, Georgia;
†††Brown Medical School, Providence, Rhode Island; and ‡‡‡Sheba
Medical Center, Tel Hhashomer, Tel Aviv University, Ramat-Gan,
Israel.
Reprints: Alex V. Levin, MD, MHSc, FRCSC, Department of Ophthalmol-
ogy, M158, The Hospital for Sick Children, 555 University Avenue,
Toronto, Ontario M5G 1X8 Canada. E-mail: alex.levin@sickids.ca.
Copyright © 2007 by Lippincott Williams & Wilkins
ISSN: 0195-7910/07/2804-0323
DOI: 10.1097/PAF.0b013e31815b4c00
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 4, December 2007
323

Page 2

and burial. Embalmers routinely engage in these procedures,
and there are existing ophthalmic pathologic protocols to
ensure “postmortem cosmesis” after larger ophthalmic and
facial autopsy excisions.
To achieve the optimal yield of useful information, it is
important that the autopsy is conducted in a goal-oriented
fashion with regard to the pathology known to occur in the
specific conditions within the differential diagnosis. Elucida-
tion of the ocular findings of inflicted childhood neurotrauma
has occurred over the last 30 years since Caffey’s first
description of what he called “the whiplash shaken infant
syndrome.”1,2Over the ensuing decades, evidence has accu-
mulated to support the concept that repetitive violent accel-
eration-deceleration forces, with or without blunt head
trauma, in the first few years of life result in a characteristic
pattern of intracranial, skeletal, and ocular injury that we will
refer to herein as the shaken baby syndrome (SBS). The
reader is referred elsewhere for reviews and additional infor-
mation beyond the scope of this article.3–11
Kiffney described retinal detachment with bilateral sub-
retinal hemorrhage in 1964.12Preretinal and intraretinal hem-
orrhage, engorgement of retinal veins, and papilledema were
reported in 1967 by Gilkes.13Mushin showed intraretinal,
subhyaloid, and small vitreous hemorrhages in 1971.14Har-
court and Hopkins described intraocular hemorrhages as well
as the ocular sequelae of such hemorrhages in 1971.15Caf-
fey’s articles in 19721and 19742mention ocular hemor-
rhages. Two fatal cases described in his 1974 article had
autopsies, but only one had a postmortem ocular examination
that revealed “bleeding in the fibrous layer and scattered
hemorrhages in the inner nuclear layer.”2As more clinicians
and pathologists became aware of the importance of ophthal-
mic manifestations in child abuse, additional observations
were presented including Purtscher retinopathy16; injuries
simulating congenital glaucoma17; retinal detachment18; white
centered retinal hemorrhages19; and paramacular retinal
folds.20,21Hemorrhagic macular retinoschisis, first described
by Greenwald and coworkers in 198622has only been re-
ported because of inflicted childhood neurotrauma with re-
petitive acceleration-deceleration forces (SBS) in children
younger than 5 years of age7except for a single case report of
a child victim of alleged head crush injury.23
The incidence of retinal hemorrhages due to inflicted
childhood neurotrauma, in particular SBS, ranges from 49%
to 100% with the higher prevalence reported from autopsy
series.7,24–29The hemorrhages may involve any or all of the
retinal layers although preretinal and intraretinal hemorrhage,
especially in the most superficial layers, is more common
than vitreous and subretinal hemorrhage.21,30–32Neverthe-
less, subretinal hemorrhage is surprisingly common in some
studies and may result in disruption of the overlying retinal
pigmented epithelium (RPE).28,31Subinternal limiting mem-
brane (ILM) blood may occur with or without macular
folds.7,22,25,28,29Vitreous hemorrhage may be due to blood
that has broken through the ILM from a traumatic retinos-
chisis.7,30Although difficult to identify in the living child,
choroidal hemorrhage or engorgement has been observed in
30% to 50% of eyes of victims of SBS at autopsy.28,30
Several series of postmortem ocular examination in
victims of abuse have been reported. Rao and coworkers
presented the ocular pathology of 14 fatally abused children
in 1988.32These eyes were compared with 16 nonabused
controls that had no ocular hemorrhages except for a single
patient with traumatic head injury. Seventy-seven pairs of
eyes, including the 14 described by Rao were studied by
Riffenburgh and Sathyavagiswaran in 1991.29,33They found
that retinal hemorrhage is the most common form of eye
damage in child abuse and is more likely to occur in infants
who have been shaken than after blunt trauma to the head.
Accordingly, retinal involvement is seen more frequently in
younger children. Munger and coworkers in 199328reported
neuropathologic and ophthalmologic observations of 12
abused children. Hemorrhages in the superficial retinal layers
and subretinal space predominated. Retinal hemorrhages
were found in all 12 cases, and intracranial hemorrhage was
found in all but one case. Cerebral edema occurred in 10
cases. The intraretinal and periretinal hemorrhages were most
prevalent at the posterior pole. Five cases had retinal folds.
There was a low incidence of optic disc edema and choroidal
hemorrhage. An autopsy study by Budenz and coworkers
highlighted the importance of ocular and optic nerve hemor-
rhages in 13 abused infants with intracranial injury.31A
prospective postmortem study reported in 1994 examined the
eyes of 169 children, both abused and nonabused.26Retinal
hemorrhages were found much more frequently in inflicted
head injury than accidental injury or natural disease. Most
other causes of retinal hemorrhages in this age group such as
leukemia and coagulation abnormalities can be readily rec-
ognized on clinical retinal examination and diagnostic inves-
tigations. In addition, retinal hemorrhages in these diseases
are less extensive, without schisis or retinal folds. Randomly
oriented retinal folds or a circumferential fold at the ora may
be fixation artifact, whereas a circumlinear paramacular fold
(often with vitreous adherent to the apex) is most commonly
a sign of SBS.
Although the first descriptions of the Battered Child
syndrome and SBS were published decades ago, there is still
confusion of terminology and reference standards. The term
“retinal hemorrhages” for example is often used in the med-
ical literature as a generic term without specification of the
hemorrhage type, severity or pattern of distribution.7Even
though hemorrhages most often involve the posterior pole,
extension to the ora serrata may be of particular signifi-
cance.7,24,28,30Before sectioning of the eye for histology and
after the eyes have been fixed and sectioned it is advanta-
geous to photograph and document the gross distribution
pattern of the hemorrhages.7,34One group proposed a grading
scheme based on microscope grids24which may be useful in
quantifying the amount of hemorrhage in a given histologic
preparation.
Failure to describe retinal hemorrhage in detail may
lead to statements claiming that a particular condition “can
result in retinal hemorrhages” even though the hemorrhages
Gilliland et al
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 4, December 2007
© 2007 Lippincott Williams & Wilkins
324

Page 3

may look dramatically different from the ones seen in in-
flicted childhood neurotrauma with repeated acceleration-
deceleration forces with or without impact (SBS). The de-
scription of ocular findings in the published literature was
often vague or absent as was the case even in Kempe’s 1962
landmark article defining the Battered Child syndrome.35One
child suffered from “a defect in vision.” No description of the
ocular injuries was given. Accurate, uniform descriptions of
postmortem ocular findings could be helpful in resolving such
uncertainties. This will be facilitated by standardization of the
forensic postmortem ocular protocol, as has been proposed in
the United Kingdom for ocular necropsy in general36and the
forensic autopsy.37
Appreciation of the importance of posterior optic nerve
and orbital findings in shaken babies has been a more recent
development. In 1986, Lambert and coworkers were the first
to describe optic nerve sheath hemorrhages which were
subarachnoid and subdural.38Optic nerve sheath hemorrhage
occurs in 65% to 100% of victims.25,26,28,31,32Elner and
coworkers25examined the eyes of 10 abused children. This
was the first report describing intrascleral hemorrhage at the
optic nerve-globe junction. Other authors have suggested
that intrascleral hemorrhage in SBS may be caused by
acceleration-deceleration trauma which results in tears of
the intrascleral ciliary vessels forming the circle of Zinn-
Haller.25–26,32,39,40Immunohistochemical staining studies of
postmortem specimens from abused children with markers
such as beta-amyloid precursor protein also identify this site
as a prime area for optic nerve damage.41,42There seems to
be a positive correlation between the finding of retinal hem-
orrhages, intrascleral hemorrhage at the optic nerve-sclera
junction and retrobulbar optic nerve sheath hemorrhage.26,31
The most common site for optic nerve sheath hemorrhage is
the immediate retrobulbar portion of the optic nerve,9with
most hemorrhage located in the subdural space.28,31Blood
can be found in the orbital optic nerve sheath despite the
absence of blood in the canalicular portion of the optic nerve
sheath.26,31,32,38
Other reported orbital findings include the presence of
intradural hemorrhage (within the dural tissue proper) of the
optic nerve sheath, and hemorrhage into the orbital fat, and
sheaths of the orbital portion of cranial nerves and extraocular
muscles.7,28,31Most recently, Marshall and coworkers pre-
sented a series of postmortem ocular examinations which
underscores the value of evaluation of abnormalities in the
orbit as well as the globe.43In a study of orbit pathology of
abused and accidentally head injured children it was found
that the combination of optic nerve sheath and orbital hem-
orrhage was significantly (P ? 0.01) more common in abu-
sive head trauma.44In inflicted neurotrauma the hemorrhage
was significantly more common in the subdural, subarach-
noid, intradural and levels of the optic nerve sheath and
occupied a larger portion along the length of the orbit.
Hemorrhage was also observed significantly more often in the
orbital fat, extraocular muscle sheaths, and cranial nerve
sheaths. Orbital injury was observed in accidental trauma
only when there was direct trauma to the orbit or severe
multiple repetitive acceleration-deceleration injury. When or-
bital hemorrhage did occur in accidentally injured children
without direct orbital fracture, the amount of hemorrhage in
the orbit and dura was minimal, being visible only histolog-
ically. Extraocular muscle hemorrhage occurred only in the
abused children.
These guidelines incorporate a rigorous reproducible
“best practice” protocol for postmortem ocular examination
in sudden unexplained death in children younger than 5 years
and in those where physical child abuse, in particular the
SBS, is suspected. Future research, such as the possible use of
staining for beta-amyloid precursor protein,41may demand
additional investigations and revision of this protocol. The
dating of hemorrhagic ocular injuries using hemosiderin as-
sessment remains somewhat controversial and imprecise and
is therefore not addressed here. The goal of this proposed
protocol is to facilitate uniform description of postmortem
ocular and orbital findings that can serve as a basis for
scientific investigation and comparison, and help to ensure
that all potentially useful findings are considered in arriving
at a diagnosis.
PROTOCOL
1. Review the enabling legislation in your jurisdiction for
authority to remove eyes at autopsy.
1.1. If child abuse is suspected, the law is likely to
identify the authorizing authority. If SIDS is under
consideration, is also very likely that the authoriz-
ing authority will be specified.
1.2. If required, informed consent for removal, retention
and ultimate disposal of eye tissues should be
obtained from the custodial caretaker.
2. Before removing the globes and orbits observe and
document:
2.1. the lids for evidence of abrasion, laceration or
ecchymoses,
2.2. the surface of the eye for any subconjunctival
hemorrhage,
2.3. the anterior chamber for hyphema or other evidence
of trauma.
Documentation should include diagrams and color pho-
tography of the eyelids, periorbital facial structures, conjunc-
tiva, and visible portions of the globe (cornea, iris, and pupil).
This should be part of full photographic documentation of the
body, including all aspects of the head and neck.
3. After removal of the brain, remove the eyes using a
combined anterior and posterior approach.
3.1. Using small cartilage cutters or small scissors, re-
move the orbital roof completely (Fig. 1). Include
the bone overlying the optic nerve canal so that the
entire nerve can be removed intact to the point of
the intracranial transection created during the re-
moval of the brain. The orbital roof should peel
away from the underlying periosteum easily. How-
ever, this separation may be more difficult over the
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 4, December 2007Postmortem Ocular Investigation
© 2007 Lippincott Williams & Wilkins
325

Page 4

canal. The goal of this step is to leave the orbital
cone of tissues en bloc with out disruption.
3.2. Using blunt dissection, separate the periosteum that
encases the orbital contents from the lateral, medial
and inferior orbit walls. Sharp dissection will only
be required to cut the attachments of the periosteum
to the inferior orbital fissure (Fig. 2), to disinsert the
apical attachments of the extraocular muscles, to
disinsert the inferior oblique muscle attachment to
the anterior medial orbital wall, to transect the
superior oblique muscle at the trochlea, to sever the
apical entry points of the nerves and vessels enter-
ing the orbit, and to remove the intracanalicular
optic nerve.
3.3. Insert an eyelid speculum to separate the lids. Be
sure not to allow the eyelids to evert.
3.4. Approaching the globe from the front, use small
scissors and forceps to cut through the conjunctiva
for 360 degrees posterior to the outside edge of the
cornea (limbus). Use care to avoid damage to the
lids.
3.5. Using the scissors, carefully blunt dissect around
the globe between the fascia which underlies the
conjunctiva (Tenon fascia) and the sclera. The tip
of the scissors will then become visible from the
superior intracranial view of the orbit as it exerts
pressure on the periosteal cone from within. Using
a scalpel or scissors, incise the periosteum over the
tip of the scissors from the orbital side. Then extend
this incision for 360 degrees taking care not to cut
the lids from the posterior approach.
3.6. Remove the globe and orbital contents en bloc
along with the intracanalicular optic nerve (Fig. 3).
4. Fix the en bloc ocular and orbital tissues in neutral
buffered formalin. The volume of formalin should be at
least 10 times that of the removed specimen. Adequate
fixation takes 2 to 3 days. Longer fixation is not harmful,
although the sclera becomes progressively harder and
more difficult to section from processed tissue blocks.
5. Step section the entire en bloc specimen from the distal
optic nerve to within 0.5 to 0.7 cm of the posterior aspect
of the globe. Avoid disrupting the globe.
5.1. Before sectioning, it may be helpful to transillumi-
nate the globe through the cornea, taking care not to
damage its surface, and note the position of any
posterior chamber shadows (indicating extensive
hemorrhages, tumors etc.). Retroilluminate the iris
and pupil (fiber optic light on the posterior sclera)
to demonstrate lens displacement, cataracts, and iris
stromal defects.
5.2. On gross inspection of the sections, document and
color photograph hemorrhages, in particular hem-
orrhage inside and outside of the optic nerve sheath
(if present).
FIGURE 3. The final en bloc specimen of globe and orbital
contents.
FIGURE 1. Using scissors or a ronjeur (shown here) the roof
of the left orbit is removed in pieces.
FIGURE 2. As the left orbit and globe are lifted from the or-
bit en bloc the attachments of the periosteum to the inferior
orbital fissure are identified and cut with a scalpel.
Gilliland et al
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 4, December 2007
© 2007 Lippincott Williams & Wilkins
326

Page 5

5.3. After hematoxylin and eosin staining, histologic
examination should in particular note the presence
of hemorrhage in the orbital soft tissues, in the optic
nerve subdural space, in the optic nerve subarach-
noid space, within the optic nerve dura proper,
inside the sheaths of cranial nerves and extraocular
muscles, and within the extraocular muscle tissue.
6. Cut the globe slowly with a new sharp 60 mm blade or
razor blade while holding the globe firmly but not
squeezing. Look for gross evidence of blood in the
vitreous. There are 4 methods for sectioning the globe.
6.1. Method #1: Alternative Section
6.1.1. Note the position of the ora serrata by transil-
lumination and mark it on the sclera with a
pencil. The section will be just anterior to this
line in the coronal plane through the pars
plana to avoid detaching the retina.
6.1.2. One calotte will be the anterior chamber,
iris and lens. The posterior chamber includ-
ing the entire retina will be in the other
calotte.
6.2. Method #2: Coronal Section
6.2.1. Cut approximately at the equator of the
globe in the coronal plane.
6.2.2. One calotte will contain the anterior seg-
ment and the anterior aspects of the poste-
rior chamber including the ora serrata. The
remainder of the retina, including the mac-
ula, will be in the posterior calotte.
6.3. Method #3: Pupil-Optic Nerve method (Axial Sec-
tion)
6.3.1. Place the cornea facing down. Start the
section posteriorly, superior to the optic
nerve stump in the axial plane (ie, parallel to
the visual axis). The superior and inferior
aspects are identified by the long posterior
ciliary vessels, the superior oblique inser-
tion and the inferior oblique insertion. More
force may be needed to cut through the lens.
6.3.2. The macula will be in the inferior calotte.
6.4. Method #4: Pupil-Optic Nerve method (Sagittal
Section)
6.4.1. Place the cornea facing down. Start the
section posteriorly, nasal to the optic nerve
stump in the sagittal plane. Nasal (medial)
and temporal (lateral) aspects can be iden-
tified by first identifying the superior-infe-
rior orientation (see 6.3.1). More force may
be needed to cut through the lens.
6.4.2. The macula will be in the temporal calotte.
7. Photograph the entire inner surface of the eye, including
the lens and in particular the retina, in color with dis-
secting microscope (or equivalent) magnification. Plac-
ing the eye under 70% alcohol (ethanol) or normal saline
will eliminate obstructing illumination highlights during
photography. Ensure that your photographs are properly
developed (if film) and in clear focus at high resolution
before proceeding with further dissection. For this rea-
son, digital photography is preferred.
Ensure that all drawings, photographs and notes are carefully
labeled and preserved as hard copy, at all stages of the
procedure, to preserve evidence at forensic standard (some
jurisdictions require specific “forensic evidence” procedures).
8. Document gross (macroscopic) retinal observations:
8.1. The retina may have lifted from the choroid and
may commonly have artifactual folds radiating
from the optic disk or in random patterns, and
circumferentially at the ora (Lange fold). The pres-
ence of blood or exudate under the retina indicates
that the detachment is not an artifact.
8.2. A complete or partial circular fold in the posterior
pole, often with a crater-like appearance, is not
artifact and raises the possibility of traumatic para-
macular folds with or without traumatic retinoschi-
sis due to vitreous traction as seen in SBS.
8.2.1. Note presence of hemorrhage within a reti-
noschisis cavity
8.2.2. Note if there is hypopigmentation underly-
ing the folds.
8.3. Count the number of hemorrhages if present and
feasible.
8.4. Describe the hemorrhage location and patterns of
distribution of hemorrhages (eg, paravascular, peri-
papillary, posterior pole, macular (within vascular
arcades), midperipheral beyond vascular arcades,
peripheral (at ora serrata)
9. Sections of globe for microscopic examination:
9.1. Microscopic sections of the eye are usually taken in
the pupil-optic nerve plane (see 6.3 and 6.4).
9.2. The macula, and in particular the fovea, must al-
ways be examined.
9.3. The optic nerve must always be included. A pupil-
optic nerve section is necessary to adequately as-
sess optic nerve sheath hemorrhage and edema as
well as intrascleral hemorrhages at the junction of
the optic nerve to the sclera.
9.4. If only a few hemorrhages are present be sure one
or more are included in the plane of section.
9.5. If traumatic retinoschisis (eg, posterior pole circular
folds) is suspected, the lesion(s) should be included
in the sections.
9.6. Sections of the anterior chamber are also required
and will be included automatically if a pupil-optic
nerve method is used. If another method of globe
division is used (see 6.1 and 6.2) then separate
sections of the calotte containing the anterior cham-
ber are necessary.
10. Histologic Examination
10.1. Stain with hematoxylin and eosin. Other stains
(eg, Prussian Blue for iron in hemosiderin) are
often useful indicators, and may be performed for
The American Journal of Forensic Medicine and Pathology • Volume 28, Number 4, December 2007 Postmortem Ocular Investigation
© 2007 Lippincott Williams & Wilkins
327