British Journal of Anaesthesia 1996; 77: 310

Mechanisms of cell injury and death

J. P. COBB, R. S. HOTCHKISS, I. E. KARL

AND

Control of the rate of cell death relative to the rate of

cell division maintains organ integrity and physiological homeostasis. Cell death is valuable for the
organism because it removes terminally injured or
unwanted cells that utilize valuable substrates and
nutrients. Likewise, cell death also has value for the
species, as it provides a mechanism for eliminating
terminally injured individuals who consume necessary societal resources or harbour toxic pathogens.
Recent advances in cellular biology have contributed
substantially to our understanding of the processes
of cell injury and death, and have provided the
molecular tools necessary to control it. This paper
reviews cell injury and adaptation; mechanisms of
cell death; the roles of cell injury and death in the
pathophysiology of organ dysfunction; and implications for prevention and therapy of multiple organ
dysfunction syndrome (MODS).

Cell injury and adaptation

Cell injury occurs as a result of physical, chemical or
biological insults or as a result of vital substrate
deficiency (table 1). The cellular response to these
injuries is adaptive, designed to restore homeostasis
and protect the cell from further injury. Although
characteristic changes in gene transcription occur, it
is not the relative amount but the pattern of
transcription that changes, with emphasis directed
towards transcription of vital genes. The
responses induced by cellular injury fall into four
main patterns: the ischaemic/anoxic, oxidative, heat
shock and acute phase responses. These are reviewed
briefly below.
ISCHAEMIC/ANOXIC STRESS RESPONSE

[13]

Because of the unusual high efficiency of the

cardiopulmonary system to transport oxygen, cells in
higher animals have not developed elaborate cellular
pathways to adapt to hypoxia and are thus relatively
sensitive to ischaemia [13, 18]. As a consequence,
lack of oxygen dramatically increases the need for
anaerobic glycolysis to maintain intracellular ATP

(Br. J. Anaesth. 1996; 77: 310)

Key words
Cells, apoptosis. Cells, necrosis. Cells, death. Complications,
sepsis.

T. G. BUCHMAN

reserves. The associated changes in gene expression

include decreased total protein synthesis, induction
of hypoxia-associated proteins (e.g. glyceraldehyde3-phosphate dehydrogenase, a glycolytic enzyme),
induction of the heat shock response (see below), and
induction of glucose-regulated proteins. If these
changes are inadequate to prevent ATP depletion,
membrane ion pumps fail and membrane integrity is
lost. Increased intracellular Ca2; occurs and a variety
of degradative processes are initiated, leading to
cytoplasmic swelling and eventual cell death.
OXIDATIVE STRESS RESPONSE

[9]

Agents that provoke oxidative injury include products of oxidative metabolism (especially from the
mitochondria) and those released from activated
phagocytes. Collectively, they are called reactive
oxygen species (ROS), and include superoxide, hydrogen peroxide, hydroxyl radical (the generation of
which depends on availability of ferrous ion and
superoxide) and nitric oxide. Iron, which is essential
for DNA synthesis and oxidative metabolism, also
catalyses the Fenton and Haber-Weiss reactions,
converting superoxide to molecular oxygen and
hydrogen peroxide to the hydroxyl radical. Under
normal conditions, free iron is compartmentalized
by the protein ferritin, which protects the cell by
keeping iron in a non-reactive crystallized core as
ferric ion (Fe3;). The results of the oxidative
reactions of ROS include de-energization of mitochondria and loss of energy stores, peroxidation and
disruption of lipid membranes, and direct DNA
damage.
The adaptive response of the cell to oxidative
stress includes both enzymatic and non-enzymatic
capacities and induction of oxidative stress response
proteins [9]. Superoxide dismutase reduces superoxide to molecular oxygen and hydrogen peroxide;
catalase catalyses the conversion of this hydrogen
peroxide to oxygen and water. Reduced glutathione
(which reacts with hydrogen peroxide in the presence
of glutathione peroxidase to form water and oxidized
glutathione), vitamin C and vitamin E constitute the

J. PERREN COBB*, MD, TIMOTHY G. BUCHMAN, PHD, MD

(Department of Surgery); RICHARD S. HOTCHKISS, MD (Department of Anesthesia); IRENE KARL, PHD (Department of Medicine);
Washington University School of Medicine, St Louis, MO, USA.
*Address for correspondence: Campus Box 8109, Department
of Surgery, 660 South Euclid Ave, St Louis, MO 63110-1093,
USA.

British Journal of Anaesthesia

Table 1 Mechanisms of cell injury

(1) Physical
(a) ionizing radiation
(b) temperature
(c) mechanical trauma
(2) Chemical
(a) drugs
(b) poisons
(3) Biological
(a) enzymes
(b) cytokines
(c) viral infection
(d) cell-mediated
(4) Critical substrate deficiency
(a) oxygen
(b) glucose

major intracellular reducing (non-enzymatic) agents.

Several proteins are induced under oxidative stress,
the most important of which may be metallothionein,
which binds not only heavy metals but also ROS via
the free sulphydryl groups of its many cysteine
residues. ROS also activate the transcriptional
factors AP-1 and NFB, both of which promote
transcription of cytokines and have been associated
with induction of apoptotic cell death (see below). If
these adaptive responses are inadequate to prevent
depletion of cellular glutathione, then cell protein
thiol groups become the remaining reducing agents,
leading to loss of critical enzymatic function and cell
death.

HEAT SHOCK RESPONSE

[10, 20]

The highly conserved cellular response to heat,

known as the heat shock response, is associated with
induction of heat shock proteins (HSP). The most
extensively studied are the HSP 70 family, which
includes stress-induced HSP 72. Interestingly, HSP
also are induced by a variety of other cellular insults,
including ischaemia/reperfusion, oxidative stress,
exposure to heavy metals (e.g. arsenite) and infection. For this reason, the heat shock response is
called simply the stress response, and HSP are
known as stress proteins. Control of HSP transcription is mediated by heat shock element activation in the heat shock protein promoter region.
HSP aid in the proper folding of newly translated
proteins, and are referred to as molecular chaperones. Increased production of HSP is believed to
provide an adaptive advantage to stressed cells by
increasing the fidelity of protein synthesis and aiding
refolding of damaged or denatured proteins.

ACUTE PHASE RESPONSE

[26]

Although acute phase response can refer to the

response of any tissue to injury, its use is commonly
restricted to the dramatic change in the pattern of
hepatocyte protein synthesis. The most important
stimuli are interleukin 1 (IL-1), tumour necrosis
factor (TNF), and IL-6, products of macrophage/
monocyte activation. The hepatocyte response to
these cytokines is an outpouring of plasma proteins

that function to maintain the homeostasis of the

organism. They include C-reactive protein, fibrinogen, complement, the metal binding proteins haptoglobin and ferritin, and plasminogen activator inhibitor and many others. The acute phase response
(activation) of endothelial cells includes increased
surface expression of adhesion, selectin and integrin
molecules that facilitate leucocyte adhesion and
release of IL-1, IL-6, IL-8 and platelet activating
factor (PAF).

Mechanisms of cell death

If the genetic and metabolic adaptive responses
described above are inadequate for a given injury,
the cell will die. Increased interest in the mechanisms
responsible for cell death, however, has made it clear
that morphological classification of cell death
requires revision. At least two types of cell death
have been described. In the first, cell damage is
manifested by cytoplasmic swelling, plasma membrane blebbing, dilation of the endoplasmic reticulum and mitochondria, dissolution of chromatin and,
finally, interruption of membrane integrity. The
other, less frequently observed type of cell death was
described by Kerr [16] and called shrinkage necrosis.
It is manifested by cytoplasmic shrinkage, larger
plasma membrane buds and nuclear chromatin
condensation. Both types of cell death result in the
generation of necrotic debris that is engulfed by
phagocytic cells.

TERMINOLOGY

The morphological changes described above are not

restrictive. Nuclear condensation and mitochondrial
swelling, for example, have been observed in both
swelling and shrinkage types of cell death. There is
also clear evidence of significant overlap between
them at the molecular level (see below) [11, 14, 18].
Moreover, there is considerable imprecision in the
literature regarding the terminology of cell death.
Historically, necrosis has been the term used to refer
to cell death in general. The term apoptosis (coined
by Kerr, Wyllie, and Currie from the Greek word
meaning dropping off , as in leaves dropping from
a tree [17]) is now used specifically to describe death
manifested by cell shrinkage. Apoptosis is responsible for the ordered, normal cell death of intestinal
epithelial cells and blood neutrophils.
Programmed cell death, which refers to the ordered
death of cells during embryogenesis, however, is also
characterized by cell shrinkage. Apoptosis and
programmed cell death are frequently, but incorrectly, used interchangeably. The question of
what to call non-apoptotic cell death has been
problematic. Necrosis, a leading contender, is also
used to refer to the process of removal of the cellular
remains of apoptosis (so-called post-apoptotic
necrosis [18]) [11, 32]. This degree of imprecision
led Farber recently to remark that there is no field
of basic cell biology and cell pathology that is more
confusing and unintelligible than is the area of
apoptosis versus necrosis [11]. Consequently, several

Figure 1 Electron micrographs of murine thymocytes from normal (A, sham laparotomy) and septic (B, caecal ligation and puncture) mice. Note the differences in the pattern and
distribution of chromatin. Compared with controls (A), thymocytes from septic animals (B) demonstrate nuclear condensation and cell shrinkage, consistent with apoptosis.

Mechanisms of cell injury and death

5

British Journal of Anaesthesia

Figure 2 Micrographs (
360 magnification) of murine thymocytes from normal (A, sham laparotomy) and septic
(B, caecal ligation and puncture) mice photographed using the fluorescent TUNEL (terminal deoxynucleotidyl
transferase-mediated dUTP nick end-labelling) technique. Cells are permeabilized and treated with fluorescein
labelled dUTP and terminal deoxynucleotide transferase. Note that in the control mice, all nuclei appear dark (A).
In contrast, nuclei of septic mice (B) are fluorescent, consistent with DNA strand breaks and apoptosis.

authors have reviewed this subject in an attempt to

standardize terminology and separate the process
from the morphology of cell death [11, 14, 15, 18].
PROCESS OF CELL DEATH

From a pathological standpoint, it is important to

distinguish between the process of cells dying per se
from the changes that occur in cells after death [11,
18]. Majno and Joris [18] have characterized the
process of cell death as (1) cell death by ischaemia, (2)
cell death by accident, or (3) cell death by suicide.
Accidental cell death is used to describe death as a
result of an external agent or toxin. Both cell death
secondary to ischaemia and accident are usually
characterized morphologically by cell swelling. In
this type of cell death, interference with ATP
generation secondary to hypoxia or toxin-induced
increases in plasma membrane permeability produce
membrane failure. Reversible consequences of injury
include cytoplasmic blebbing, nuclear chromatin
condensation, dilation of the endoplasmic reticulum,
condensation then swelling of mitochondria and
activation of stress response genetic programmes
(described above). Continued injury results in interruptions of membrane integrity, dissolution of
chromatin, and calcifications within mitochondria,
all of which are signs of irreversible cell injury and
death. Majno and Joris [18] have suggested the term
oncosis, from the Greek word meaning swelling,
for this type of cell death. Necrosis is the morphological term used for the cellular debris remaining after either oncosis or apoptosis [18].

Death by suicide refers to the process of cell death

in which injury activates a highly conserved programme of suicide genes that engineer death of
the cell. This process typically results morphologically in apoptosis. Interest in cell death by
suicide (apoptosis) has gained considerable momentum recently as the disorders associated with apoptotic cell death have expanded to include cancer,
autoimmunity, inflammation, infection, AIDS,
neurodegeneration and myelodysplasia [31]. The
molecular trigger responsible for induction of apoptosis is incompletely defined, but appears to be
present in all mammalian cells at all times and is
conserved across species. Not surprisingly, induction
of apoptosis is tightly controlled. Several regulatory
genes have been identified, including the Bcl-2
family, p53 and Fas [21, 25]. The molecular machinery responsible (the executioner) includes a
cysteine protease (CPP-32 in mammals) that may
inactivate DNA repair enzymes [19, 23, 35]. The
cellular changes characteristic of apoptosis include a
flip of phosphatidylserine from the inside to the
outside surface of the cell membrane [33], cytoplasmic shrinkage, little or no swelling of organelles
(mitochondria), membrane budding and fragmentation that include mitochondria or nuclear fragments, and chromatin condensation consisting of
DNA fragments. These membrane changes lead to
rapid recognition and cell elimination by neighbouring phagocytes, which may make apoptotic cells
difficult to locate using conventional microscopic
techniques. The techniques that are used to detect
apoptosis include light and electron microscopy (fig.

Mechanisms of cell injury and death

Figure 3 Normal cultured porcine aortic endothelial cells (A,
360 magnification) stained with nuclear binding
dye Hoechst 33342. The nuclei are smooth and ovoid. Nuclei from cells challenged with endotoxin followed by
arsenite are compact and fragmented (apoptotic bodies, B,
360 magnification).

Table 2 Comparison of characteristics of oncosis and apoptosis

Oncosisdeath by swelling

Apoptosisdeath by shrinkage

(1) Swelling of organelles and cytoplasm

(2) Small dense chromatin clumps which are not sharply
defined
(3) Swelling and eventual disintegration of organelles (e.g.
mitochondria and endoplasmic reticulum)
(4) Early focal disruption of the plasma membrane with
blebbing

(1) Shrinkage of cytoplasm

(2) Large, dense, often crescent-shaped aggregates of
chromatin and nuclear fragmentation
(3) Organelles maintain structural integrity

(5) DNA agarose gel electrophoresis demonstrates smear

pattern indicative of randomized breakdown
(6) Cells rupture and contents are released causing
inflammatory response

1), end-labelling of DNA fragments (fig. 2), DNA

agarose gel electrophoresis and binding of nuclearspecific dyes (fig. 3).
Despite the relatively clear distinction which can
be made morphologically (oncosis versus apoptosis
is compared in table 2), examination at the
molecular level demonstrates considerable overlap.
For example, both oncosis and apoptosis have been
associated with increases in intracellular Ca2;, which
activate a number of enzymes including phospholipases, endonucleases, proteases and protein kinases.
In addition, the adaptive response to ischaemic,
inflammatory, oxidant and heat-induced stress involve changes in gene expression that are shared by
both types of cell death. Further, the same type of

(4) Plasma membrane maintains integrity early; later stage

characterized by budding of membrane, frequently
containing organelles
(5) DNA agarose gel electrophoresis demonstrates ladder
pattern of discrete internucleosomal breakdown
(6) Cells are either ingested whole by phagocytic cells, or they
break into membrane-bound fragments (apoptotic bodies)
which then are ingested

injury can result in cell death by either mode.

Indeed, Kerrs original description of apoptosis
(shrinkage necrosis) was based upon differentiation of shrunken necrotic cells from the more
common swollen necrotic cells in ischaemic liver
tissue [16].

Pathophysiology of organ dysfunction

Studies of cellular responses in vivo indicate that
injury activates the programmes of stress gene
expression reviewed above. For example, the hepatocellular response to ischaemic injury in a porcine
model of MODS includes activation of the acute
phase, oxidative, and heat shock responses [7]. In

8
contrast, neighbouring hepatic endothelial cells in
this model did not stain for heat shock gene
expression (in situ hybridization), suggesting that
there are cell-specific differences in the genetic
response to similar injury. A subsequent investigation conducted in vitro under more controlled
conditions explored the time-course and interaction
of these cellular stress responses. Human hepatoblastoma (HepG2) cell lines were treated with
inducers of the acute phase response (IL-1 and
TNF) and heat shock responses separately and in
combination. The results indicated that shockinduced expression of these responses were distinct,
exclusive and prioritized [6]. Specifically, the heat
shock stress response, the most primitive and highly
conserved of stress gene responses, was capable of
extinguishing expression of other genetic programmes [5, 28]. These data suggest that hepatocyte
injury of sufficient severity to trigger the heat shock
response can prevent production of system-stabilizing acute phase proteins and other adaptive
responses. The livers response to such a systemic
injury, from the hosts perspective, would appear to
be inadequate and indicative of failure. Hepatic
dysfunction in this model of MODS, thus, would
result from execution of primitive, protective genetic
responses that give individual cells priority over the
organism.
The sequence of stress gene programme activation
also appears to be important. In one in vitro model,
the endothelial cell response to arsenite, an inducer
of the heat shock response, and to endotoxin, an
inducer of the acute phase response, were studied.
The results indicated that induction of the heat
shock response could protect against endotoxininduced cytotoxicity, but reversing the order
endotoxin challenge followed by induction of the
heat shock responseincreased cell death by apoptosis [5]. The effect of the heat shock response was
not the result of a detrimental effect of the heat shock
proteins themselves, but rather to the heat shockinduced inhibition of further protein synthesis [5].
An important hypothesis drawn from these data is
that two independent programmes of stress gene
expression can interact, depending on the sequence,
to either protect the cell or trigger cell death by
suicide (apoptosis). This may explain the absence of
endothelial cells expressing heat shock proteins after
shock in vivo ([7], discussed above), as endothelial
cell apoptosis may have resulted in their subsequent
removal by phagocytes before in situ staining. Thus,
endothelial cell dysfunction and death in this model
result from the adverse interaction of two usually
beneficial patterns of stress responses. The mechanism responsible may involve alterations in the
cellular redox potential and changes in nuclear
transcription factor activity [unpublished observations]. For example, in the cultured endothelial cell
model, reducing agents such as N-acetyl-L-cysteine
and dithiothreitol block induction of endotoxin/
arsenite-induced endothelial cell apoptosis, implicating ROS in the induction of apoptotic cell death
[1, 2]. This raises the possibility that the adaptive
response produced by oxidative stress could modulate the interaction of the acute phase and heat shock

British Journal of Anaesthesia

responses, and may participate in determining the
fate of the cell.
Apoptosis may also contribute to failure of organs
in other settings, in particular, septic shock. Injection
of Gram-negative or Gram-positive bacteria [34],
endotoxin [34, 36], or caecal ligation and puncture
[4] induced apoptosis in rodent thymocytes. Zhang
and colleagues [36] reported that mice injected with
lipopolysaccharides had apoptosis not only in the
thymus, but also in the spleen, bone marrow and
endothelial cells. Norimatsu and associates injected
swine intravenously with 0.5 mg kg91 Escherichia coli
endotoxin and noted that cells in both thymus and
mesenteric lymph nodes underwent apoptosis [24].
In a recent report, Rogers and colleagues demonstrated apoptosis in the liver of mice infected with
Listeria monocytogenes [27]. Apoptosis has also been
reported in rabbit heart [12] and rat kidney [29] after
ischaemia/reperfusion injury. Collectively, these
investigations suggest that apoptotic cell death may
be a principal mechanism responsible for organ
dysfunction and death as a consequence of shockinduced injury.

Conclusionsfrom basic science to therapy

and prevention
Cell injury occurs as a result of physical, chemical or
biological insults or from vital substrate deficiency.
These insults induce expression of adaptive stress
response gene programmes that include the
ischaemic/hypoxic stress, oxidative stress, heat
shock and acute phase responses. If the severity of
the injury exceeds the ability of the cell to adapt,
then cell death occurs. The processes of cell death
include cell death by ischaemia, accident or suicide.
Morphologically, ischaemic and accidental deaths
are usually manifested by cytoplasmic swelling
(oncosis) and suicidal cell death by cytoplasmic
shrinkage (apoptosis). However, this terminology
lacks precision, as some accidental causes of cell
death, such as heat, can also induce apoptosis [18].
The molecular machinery responsible for cell death
and the boundaries between apoptosis and other
modes of cell death remain to be determined [14].
The sequence and interaction of the acute phase,
heat shock and oxidative stress responses appear to
be a determinant of the fate of the cell. The pattern
of cell responses is injury- and tissue-dependent and
may help to explain differences in the course and
nature of organ failure. Specifically, it appears that
stimuli sufficient to trigger the heat shock response
can precipitate cell (and organ) failure by blocking
initiation of other programmes of stress gene expression (e.g. the system stabilizing acute phase
response of hepatocytes). In other cells (e.g. the
endothelium), the sequence of acute phase followed
by heat shock response may result directly in
apoptotic cell death. The function of shock-induced
apoptosis is not known, but it may contribute to
control of immune cell proliferation in thymus,
spleen, bone marrow, and lymph nodes and modulation of the inflammatory response.
MODS is the result of cellular responses to a
stimulus of either enormous magnitude or distri-

Mechanisms of cell injury and death

bution. The stimulus itself frequently does not result
directly in cell death, but rather the magnitude of the
appropriate response from stimulated cells collectively is injurious to the organism. For instance,
small doses of i.v. lipopolysaccharide (endotoxin) in
humans are not directly cytotoxic, but lead to the
release of the intercellular mediators, such as cytokines, that amplify the stimulus and trigger organ
dysfunction [30]. Unfortunately, attempts to attenuate the magnitude of this response by attacking
the circulating mediators themselves (e.g. antiendotoxin and anti-cytokine therapies) have failed to
improve patient survival in randomized, prospective
trials [22]. These negative data forced a broad reexamination of the treatment strategies for the
systemic inflammatory response syndrome (SIRS)
and MODS and a change in focus from extracellular
signals (i.e. the mediators) to the intracellular
responses to these signals [8]. Recent symposia have
addressed this important issue (e.g. The Future of
Sepsis Research, August 1995, Bethesda, MD,
USA, sponsored by the American College of Chest
Physicians, National Institute of Allergy and Infectious Disease, National Heart, Lung, and Blood
Institute and National Institutes of Health, USA).

A NEW THERAPEUTIC STRATEGY: MODULATION OF

CELLULAR RESPONSES TO INJURY

In summary, the data presented above are consistent

with a new hypothesis on the cellular origin of
MODS, one that has emerged as a result of a change
in focus from the extracellular to the intracellular
response to injury. They suggest that there are
distinct, exclusive, and prioritized genetic programmes expressed in response to cell injury that are
specific to cell type and injury. The effects of these
stress response gene programmes are usually cytoprotective, but when activated in sequence, they can
precipitate apoptotic cell death. Recent success in
modulating induction of apoptosis with reducing
agents [1], anti-oxidants [2], inhibitors of protein
synthesis [5], anti-TNF antibody [34] and steroid
antagonists [3] suggests that pharmacological control
in the clinical setting is possible. A complete
understanding of the type, sequence, interaction and
impact of stress gene responses to injury will form
the basis for novel gene-directed therapy for the
treatment and prevention of MODS.

References
1. Abello PA, Fidler SA, Buchman TG. Thiol reducing agents
modulate induced apoptosis in porcine endothelial cells.
Shock 1994; 2: 7983.
2. Abello PA, Fidler SA, Bulkley GB, Buchman TG. Antioxidants modulate induction of programmed endothelial cell
death (apoptosis) by endotoxin. Archives of Surgery 1994;
129: 134140.
3. Ayala A, Herdon CD, Lehman DL, DeMaso CM, Ayala CA,
Chaudry IH. The induction of accelerated thymic programmed cell death during polymicrobial sepsis: control by
corticosteroids but not tumour necrosis factor. Shock 1995; 3:
259267.

9
4. Barke RA, Roy S, Chapin RB, Charboneau R. The role of
programmed cell death (apoptosis) in thymic evolution
following sepsis. Archives of Surgery 1994; 129: 12561262.
5. Buchman TG, Abello PA, Smith EH, Buckley GB. Induction
of heat shock response leads to apoptosis in endothelial cells
previously exposed to endotoxin. American Journal of Physiology 1993; 265: H165H170.
6. Buchman TG, Cabin DE. Molecular biology of circulatory
shock III. HepG2 cells demonstrate two patterns of shockinduced gene expression which are independent, exclusive,
and prioritized. Surgery 1990; 108: 902911.
7. Buchman TG, Cabin DE, Vickers S, Deutschman CS,
Delgado E, Sussman MM, Bulkley GB. Molecular biology of
circulatory shock II: Expression of four groups of hepatic
genes is enhanced following resuscitation from cardiogenic
shock. Surgery 1990; 108: 559566.
8. Buchman TG, Zehnbauer BA. Molecular biology in the
intensive care unit: a framework for interpretation. New
Horizons 1995; 3: 139145.
9. Camhi SL, Lee P, Choi AMK. The oxidative stress response.
New Horizons 1995; 3: 170182.
10. De Maio A. The heat-shock response. New Horizons 1995; 3:
198207.
11. Farber E. Programmed cell death: necrosis versus apoptosis.
Modern Pathology 1994; 7: 605609.
12. Gottlieb RA, Burleson KO, Kloner RA, Babior BM, Engler
RL. Reperfusion injury induces apoptosis in rabbit cardiomyocytes. Journal of Clinical Investigation 1994; 94: 1621
1628.
13. Graven KK, Farber HW. Hypoxia-associated proteins. New
Horizons 1995; 3: 208218.
14. Hockenbery D. Defining apoptosis. American Journal of
Pathology 1995; 146: 1619.
15. Kane AB. Redefining cell death. American Journal of
Pathology 1995; 146: 12.
16. Kerr JFR. Shrinkage necrosis: a distinct mode of cellular
death. Journal of Pathology 1971; 105: 1320.
17. Kerr JFR, Wyllie AH, Currie AR. Apoptosis: a basic
biological phenomenon with wide-ranging implications in
tissue kinetics. British Journal of Cancer 1972; 26: 239257.
18. Manjo G, Joris I. Apoptosis, oncosis and necrosis. American
Journal of Pathology 1995; 146: 315.
19. Martin SJ, Green DR. Protease activation during apoptosis:
death by a thousand cuts? Cell 1995; 82: 349352.
20. Minowada G, Welch WJ. Clinical implications of the stress
response. Journal of Clinical Investigation 1995; 95: 312.
21. Nagata S, Golstein P. The Fas death factor. Science 1995;
267: 14491456.
22. Natanson C, Hoffinan WD, Suffredini AF, Eichacker PQ,
Danner RL. Selected treatment strategies for septic shock
based on proposed mechanisms of pathogenesis. Annals of
Internal Medicine 1994; 120: 771783.
23. Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP,
Ding CK, Gallant M, Gareau Y, Griffin PR, Labelle M,
Lazebnik YA, Munday NA, Raju SM, Smulson ME, Yamen
T, Yu VL, Miller DK. Identification and inhibition of the
ICE/CED-3 protease necessary for mammalian apoptosis.
Nature 1995; 376: 3743.
24. Norimatsu M, Ono T, Aoki A, Ohishi K, Takahashi T,
Watanabe G, Taya K, Sasamoto S, Tamura Y. Lipopolysaccharide-induced apoptosis in swine lymphocytes in
vivo. Infection and Immunity 1995; 63: 112116.
25. Olvati ZN, Korsmeyer SJ. Checkpoints of dueling dimers foil
death wishes. Cell 1995; 79: 189192.
26. Pannen BHJ, Robotham JL. The acute phase response. New
Horizons 1995; 3: 183197.
27. Rogers HW, Callery MP, Deck B, Unanue ER. Listeria
monocytogenes induces apoptosis of infected hepatocytes.
Journal of Immunology 1996; 156: 679684.
28. Schoeniger LO, Reilly P, Bulkley GB, Buchman TG.
Molecular biology of circulatory shock V: Heat shock gene
expression excludes hepatic acute phase gene expression
following resuscitation from hemorrhagic shock. Surgery
1992; 112: 355363.
29. Schumer M, Colombel MC, Sawczuk IS, Gobe G, Connor J,
OToole KM, Ollsson CA, Wise GJ, Buttyan R. Morphologic, biochemical, and molecular evidence of apoptosis
during the reperfusion phase after brief periods of renal
ischemia. American Journal of Pathology 1992; 140: 831838.

10
30. Taveira Da Silva AM, Kaulbach HC, Chuidian FS, Lambert
DR, Suffredini AF, Danner RL. Brief report: shock and
multiple-organ dysfunction after self-administration of Salmonella endotoxin. New England Journal of Medicine 1993;
328: 14571460.
31. Thompson CB. Apoptosis in the pathogenesis and treatment
of disease. Science 1995; 267: 14561462.
32. Trump BF, Berezesky IK. Calcium-mediated cell injury and
cell death. FASEB Journal 1995; 9: 219228.
33. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger
C. A novel assay for apoptosis. Flow cytometric detection of
phosphatidylserine expression on early apoptotic cells using

British Journal of Anaesthesia

fluorescein labelled Annexin V. Journal of Immunological
Methods 1995; 184: 3951.
34. Wang S-D, Huang KJ, Lin YS, Lei HY. Sepsis-induced
apoptosis of the thymocytes in mice. Journal of Immunology
1994; 152: 50145023.
35. Whyte M, Evan G. Apoptosis: the last cut is the deepest.
Nature 1995; 376: 1718.
36. Zhang XM, Morikawa A, Takahashi K, Jiang GZ, Kato Y,
Suglyama T, Kawal M, Fukada M, Yokochi T. Localization
of apoptosis (programmed cell death) in mice by adminstration of lipopolysaccharide. Microbiology and Immunology
1994; 38: 669671.