British Journal of Anaesthesia 1997; 79: 214–225
Immersion, near-drowning and drowning
, M. J. T
R. C. S
“Drowning” is defined as “suffocation by
submersion, especially in water”
; it continues to be
the third most common cause of accidental death in
the general population and, for children in many
countries, the second most common cause after road
In the USA, 40% of drowning victims
are less than 4 yr old.
In Britain, the Office of
Population Census reports that child deaths from
this cause continue to be the third most common
cause of accidental death after road accidents and
“Near-drowning” is defined as “survival, at least
temporarily, after suffocation by submersion in
We would challenge this definition as most
dictionaries define “suffocation” as cessation of
breathing leading to unconsciousness or death.
However, pulmonary complications may follow
aspiration of water without cessation of breathing or
loss of consciousness. Thus “near-drowning” should
be defined as “survival, at least temporarily, after
aspiration of fluid into the lungs”. The importance of
the distinction between the two definitions is that
aspiration of fluid may lead to later pulmonary
complications, even in those without a history of loss
of consciousness, and thus care must be exercised in
the management of all patients with a history of
aspiration. Modell, Graves and Ketover reported
that in dogs, aspiration of as little as 2.2 ml/kg body
weight produced a decrease in
aP to approximately
8 kPa (60 mm Hg) within 3 min. Pulmonary
surfactant is altered locally by water in the alveoli
resulting in pulmonary shunting via either fluid-filled
(salt water) or atelectatic (fresh water) alveoli.
stated that within minutes after inhalation of
as little as 2.5 ml/kg body weight, the normal
intrapulmonary shunt of approximately 10% may
increase to as much as 75%. Even victims who are
conscious, alert and outwardly clinically normal after
a near-drowning incident may take several days to
revert to pre-immersion values.
After aspiration of
either fresh or salt water there may be delayed
outpouring of fluid into the alveoli, secondary to
pulmonary parenchymal damage with transuda-
tion of protein rich fluid, so called “secondary
(Br. J. Anaesth. 1997; 79: 214–225).
Drowning. Hypothermia. Heart, cardiopulmonary bypass.
Heart, resuscitation. Immersion.
with consequent impairment of gas
cited several cases who
appeared normal on assessment in the emergency
department, even with normal chest x-rays, but who
developed fulminant pulmonary oedema as long as
12 h after the near-drowning incident.
While statistics on the incidence of drowning
deaths are available for most nations, there are no
precise data for near-drowning incidents, although
in the USA they are estimated to be 500–600 times
more common than their fatal counterpart.
Britain, a comprehensive hospital survey for the
years 1988–1989 by Kemp and Sibert
drowning and near-drowning incidents in children
reported 142 deaths and 188 survivors.
One of the great tragedies of drowning or near-
drowning is that the victims, frequently young, are
generally in good health before the unexpected event
which resulted in death or brain damage in a number
of survivors. In recent years increased public aware-
ness of the general principles of basic life support
and cardiopulmonary resuscitation (CPR) has
resulted in many more near-drowning victims
arriving in hospital in a state capable of being
resuscitated. The many contemporary reports in the
literature of apparently hopeless cases being success-
fully resuscitated, particularly after cold water
11 20–22 38 45 53 54 59 85 86 102
has led to
protracted resuscitative efforts by hospital personnel.
Drowning and near-drowning have been the
subject of many reviews.
15 64 66 74 78 88
tion and the subsequent respiratory management of
the near-drowning patient is now well established
and largely non-controversial, the remaining major
therapeutic challenge is limitation of brain damage
in survivors with its associated consequent human
and economic costs. Although the success stories
tend to be reported, in the absence of full statistical
information it is difficult to obtain an overall picture
of the ratio of those who have been resuscitated and
left with or without residual brain damage.
The criteria used to identify those with a poor
prognosis needs to be refined. Some centres, based
on retrospective analysis, have reported their results
of various treatment regimens with suggested predic-
tive criteria for recovery without severe brain
34 53 68 74
Regrettably, these criteria are often
, M. J. T
Robens Institute, University of Surrey, Guildford, Surrey GU2
R. C. S
Royal Navy, RH Haslar, Gosport, Hants PO12 2AA.
Immersion, near-drowning and drowning 215
conflicting and thus of little real value other than to
act as a reminder that long-term complications occur
and caution should be exercised in discussing prog-
nosis with relatives in the early stages. One of the
problems appears to be that, although the protective
role of hypothermia is now generally recognized, it is
not always possible to determine from deep body
temperature on arrival at hospital what the cerebral
temperature was before severe hypoxia developed.
This review examines the physiological considera-
tions of accidental immersion and discusses the
factors which may influence outcome. The recom-
mended management of such victims is also
reviewed, with particular emphasis on the role of
The physiological effects of immersion in
thermoneutral water have been reviewed previously
In sum, changes which occur are
largely a consequence of the effects of the external
hydrostatic pressure on the resilient body tissues and
gas-containing body cavities during head-out
immersion. This results in physiological alterations
in the cardiovascular and respiratory systems, and in
the gastrointestinal tract. The increase in venous
return causes a 32–66% increase in cardiac output,
increase in right to left shunting as a result of a
combination of pulmonary pooling, cephalic move-
ment of the diaphragm and, to a lesser extent, minor
compression of the chest wall; overall there is a 65%
increase in the work of breathing. The central shift
in blood volume is sensed by the body as
hypervolaemia and results in diuresis, natriuresis
From a practical viewpoint there is little evidence
that these changes cause respiratory or circulatory
embarrassment in a healthy individual.
during rescue, after protracted immersion in warm
water, or relative shorter immersion in cold water
when hypothermia may be present, circulatory
collapse and cardiac arrest may occur. This
phenomenon, which has been termed “circum-
rescue collapse” is considered to be outside the
scope of this article and has been reviewed
The increase in right atrial pressure, resulting from
reflex peripheral cold vasoconstriction after
immersion in cold water, is additive to the hydro-
static pressure effect described above and enhances
urinary output by as much as one-third.
of more importance to the accidental immersion
victim are a group of cardiorespiratory reflex
responses which can occur on sudden immersion in
cold water and last for approximately 2–3 min.
These responses, collectively known as the “cold-
shock” response, have been reviewed by Tipton.
They are initiated by a rapid decrease in skin
temperature and are probably responsible for the
majority of near-drowning incidents and drowning
deaths after accidental immersion in open waters in
the UK. Respiratory drive is enhanced on immersion
in water colder than 25 ⬚C
and is inversely related
to water temperature, reaching a maximum level
with immersion in water at 10 ⬚C.
includes an initial “gasp” of 2–3 litre in an adult, fol-
lowed by uncontrollable hyperventilation which can
result in a 10-fold increase in minute ventilation
which significantly reduces
shift in end-expiratory lung volume after cold water
immersion can result in tidal breathing within 1 litre
of total lung capacity, creating a sensation of
dyspnoea. This, coupled with the swim stroke/
respiration asynchrony caused by hyperventilation,
makes swimming very difficult, and is believed to be
one of the major causative factors in the mechanism
of cold-water swim failure.
The maximum breath-hold times of normally
clothed individuals are reduced to less than 10 s.
As a consequence, on immersion in choppy or
turbulent water where the airway may be intermit-
tently submerged, there is a significant chance of
aspiration during the first few minutes. Breath-
holding to facilitate escape from a submerged vehicle
may be equally difficult and result in entrapment
and drowning, or possibly near-drowning in
those who manage to escape or are rescued before
a lethal volume of water is aspirated, approximately
22 ml kg
Thus in healthy young people, who may be
competent swimmers but unhabituated to cold, the
initial respiratory responses to immersion in cold
water may cause short-term incapacitation which
could prevent them from relaxing and swimming a
short distance to save their lives. In turbulent water,
or if escape from a submerged vehicle is necessary,
the inability to breath-hold for more than 10–20 s
may result in aspiration and near-drowning, or even
The initial cardiovascular responses to cold
immersion include an immediate intense reflex
peripheral vasoconstriction, 42–49% increase in
heart rate and 59–100% increase in cardiac output,
with a resulting increase in arterial and venous
pressures. These responses significantly increase the
workload of the myocardium and, coupled with the
concomitant increased catecholamine response,
5 16 48
may induce undesirable cardiac arrhythmias.
Additionally, in middle-aged or elderly people
with heart disease, hypertension or vascular disease,
the initial cardiovascular responses to immersion in
cold water may precipitate a cardiovascular accident.
It is considered that this mechanism as a cause of
incapacitation resulting in drowning is probably
underestimated. The eyewitness reports of the
sudden cessation of movement in some drowning
is more suggestive of sudden circulatory
malfunction. The fact that aspiration may occur
subsequent to a cardiovascular incident may
mislead some into concluding that drowning was
the only problem requiring treatment or the cause of
Given the above, it is perhaps less surprising
to discover that 60% of the annual open water
216 British Journal of Anaesthesia
immersion deaths in the UK occur within 3 m of a
safe refuge, and two-thirds of those who die were
regarded as “good swimmers”.
“Hypothermia”, defined as a deep body temperature
is an inevitable consequence of protracted
immersion in cold water and appears to play an
important role in the successful outcome of those
who have survived prolonged submersion. The
mechanism of how deep body temperature is
reduced so rapidly in some of these victims is less
The rate of change of core temperature in an
immersed body is dependent on the inter-relation-
ship of several physical and physiological factors
such as water temperature, relative movement of
water adjacent to the skin, body surface area to mass
ratio, insulation, peripheral circulation, metabolism
or conditions which may affect any of these factors
(e.g. injury, intoxication, etc).
With continued body cooling, consciousness
becomes progressively impaired until it is lost at a
deep body temperature of approximately 30 ⬚C.
Cerebral blood flow decreases in proportion to the
reduction in metabolism,
while Stone, Donnelly
reported a 6–7% reduction for each
decrease in core temperature of 1 ⬚C; cerebral
activity is abolished at brain temperatures less than
However, these data relate to anaesthetized
patients in whom shivering was controlled. In
victims of accidental immersion in cold water whose
airway is being maintained clear of the water by a life
jacket, shivering is pronounced with a resulting
increased production of carbon dioxide, at least in
the early stages, until body temperature decreases to
approximately 29 ⬚C. Should ventilation be impaired
however, shivering is depressed or even absent while
hypercapnia produces cerebral vasodilatation which
increases cerebral perfusion.
With further cooling, in spontaneous breathing
adults, cardiac arrest from ventricular fibrillation
(VF) may occur at a deep body temperature :28 ⬚C,
or asystole at approximately 24–26 ⬚C.
with surface cooling, when core temperatures
decrease to 35–33 ⬚C, skeletal muscle temperatures
are usually at or below 28 ⬚C when neuromuscular
performance is significantly impaired, making swim-
ming or other actions necessary for survival
Therefore, it is likely that most
immersion victims, excepting those wearing a life
jacket with a “spray hood”,
aspirate water as con-
sciousness becomes impaired (i.e. as core tempera-
ture decreases to less than 34 ⬚C) and thus well
before significant protection from cerebral hypoxia is
In spite of the ability of cold water to extract heat
from the immersed body, hypothermia per se is
unlikely to be a problem within 30 min of head-out
immersion for a healthy, clothed adult, even in water
as cold as 5 ⬚C. The normal physiological responses
to cold reduce heat transfer by mass flow (circula-
tion) to the surface from the deeper tissues and
increase shivering heat production by as much as
As a general rule, however, in spite of
these responses, during head-out immersion in
laboratory conditions, deep body temperature of the
average adult wearing outdoor clothing has fallen to
35 ⬚C after approximately 1 h in water at 5 ⬚C, 2 h in
water at 10 ⬚C, and 3–6 h in water at 15 ⬚C.
water however, because of the rich supply of blood
vessels to the scalp which do not vasoconstrict in the
cold, heat loss from the unprotected head may be
enhanced by forced convection and evaporation
thereby increasing the rate of body cooling consider-
ably. Similarly, if the head is immersed the rate of
body cooling is significantly faster.
Because of their greater surface area to mass ratio
and reduced subcutaneous fat content, cooling rates
of immersed children are also significantly faster
than those seen in adults. However, with surface
cooling alone, even in children, rates for decrease in
body temperature are too slow to cool the brain suf-
ficiently to protect it from the hypoxia encountered
in drowning asphyxia. In anaesthetized naked
infants, being surface cooled for surgery with ice
packs and ice cold water, rectal temperatures
decreased by as little as 2.5 ⬚C in the first 10 min of
the procedure, and it took 32.5 min for it to reduce
to 24–25 ⬚C.
Thus if cerebral hypothermia is the
mechanism through which some children can
survive protracted submersion without sustaining
lasting hypoxic neurological damage, an alternative
method of heat exchange to surface cooling alone
must be found.
To date it has been postulated by several
23 29 33 59 71 78 86
that the “mammalian diving
response”, either alone or in combination with
may be the explanation.
The diving response, characterized by apnoea,
generalized marked peripheral vasoconstriction and
is initiated reflexly by stimulation of
the ophthalmic division of the trigeminal nerve on
immersion of the face in cold water. It plays a
powerful role in oxygen conservation in diving
animals, enabling some to remain submerged for
periods in excess of 30 min. The response is
enhanced by both anxiety and cold water. The
magnitude of the diving response found in humans is
qualitatively similar to that found in diving
mammals, but quantitatively less marked, with the
cold-shock response predominating in the majority
of normally clothed adults.
15% of people do show a profound reaction
this percentage increases with the use of specialist
protective clothing, such as immersion “dry” suits,
which prevent rapid cooling of the cutaneous
thermal receptors of the majority of the body, but
leave those of the face exposed to the cold stimulus.
Tipton, Kelleher and Golden
have reported that in
such circumstances, where the cold-shock and
diving responses are stimulated in roughly equal pro-
portions, the resulting fluctuation in the competing
parasympathetic and sympathetic chronotropic
influences to the heart frequently (in 31 of 36
immersions) caused a variety of cardiac arrhythmias
Immersion, near-drowning and drowning 217
(predominantly supraventricular) just after the
cessation of breath-holding in cold water.
The diving response is also reported to be more
marked in infants than adults.
Gooden, in his
argued that while the diving
response alone is unlikely to explain the reported
survival of those cases of protracted submersion in
cold water, it does play a contributory role together
with hypothermia. He hypothesized that survival
from prolonged near-drowning appears to depend
on a specific interplay between the diving response
and hypothermia, resulting in a protective state of
“hypometabolism”. Peripheral vasoconstriction,
associated with both the diving response and cold,
depresses tissue metabolism and thus reduces body
oxygen requirements. Gooden postulated that
immediately on face immersion the diving response
apnoea prevents water aspiration into the lungs.
Even if water does enter the larynx he postulated that
reflex glottal spasm prevents further penetration into
the lungs. The associated bradycardia and selective
redistribution of blood conserves oxygen stores
thereby protecting the brain from hypoxia during the
initial minutes of immersion. Gradually, with surface
cooling, hypothermia develops so that by the time
oxygen stores are depleted, cerebral metabolism
has been reduced to a level which confers some
protection from hypoxia.
While to some extent this hypothesis is plausible
there are several questions that remain to be
answered. First, why is it that in most cases of pro-
longed submersion who survived, sufficient water
was aspirated to cause near-drowning? It is possible
that aspiration, when it did occur, was a relatively
terminal event when the diving response, or
laryngospasm, could no longer be maintained
because of profound hypoxia. However, from our
understanding of surface cooling, discussed above, it
is probable that profound hypoxia is present before
significant brain cooling has occurred in this circum-
stance. Second, does the cold-shock response over-
ride the diving response in children as it does in
The only evidence in the literature
relating to children in this context is that of Ramey,
Ramey and Hayward
who found that children
could not hold their breath long enough for a diving
response to occur; in their experiment however, the
temperature of the water was 29 ⬚C, above that
normally associated with the initiation of the most
powerful diving response. Skin temperature of
children already playing in water is at or in the region
of the water temperature and thus, if they become
accidentally submerged, should not exhibit a cold-
shock response. However, those children who fall
into water without prior wetting should display a
cold-shock response. Finally, another factor mitigat-
ing against the “hypometabolism” hypothesis is the
large discrepancy in the rates of cooling reported
(varying from 10 ⬚C in 10 min in ice water
3.5 ⬚C in 6 min in a young adult
). Such rates of
cooling are not compatible with either surface cool-
ing alone or in combination with a diving response.
Thus while not ruling out entirely the participa-
tion of the diving response in the early stages of some
of these remarkable survival stories, it is considered
that the rapid rates of cerebral cooling encountered
may be largely dependent on pulmonary heat
exchange after aspiration.
After submersion, the victim may attempt to breath-
hold for as long as possible before neurogenic
afferents from the respiratory muscles or cutaneous
thermoreceptors build to an intolerable level, forcing
the individual to take a breath.
It has been shown
that voluntary breath-holding can be extended by
movements of the respiratory muscles, and swallow-
ing, against a closed glottis.
This may be the
explanation for the large amounts of water reported
to have been found in the stomach of some drowning
11 27 67 78
but not in others
(i.e. ingestion of
water is only likely to be found in those who
attempted to extend voluntary breath-holding).
Depending on the volume and temperature of the
water swallowed, heat exchange between the blood
and gastric wall could contribute significantly to
blood cooling, especially in children. When
continued breath-holding is impossible, aspiration of
water into the airways and alveoli occurs with
inspiration and possibly reflex coughing, provided
reflex laryngeal spasm does not supervene.
Regardless of the method of entry of cold water
into the “core” of the body, be it lungs, stomach, or
both, it precipitates a decrease in deep body
temperature. The magnitude of this decrease is
directly proportional to the volume of water
aspirated, its temperature differential with the body
and state of the circulation. However, to protect the
brain from hypoxic damage it would be necessary to
cool the brain very rapidly, at least 7 ⬚C in 10 min.
This doubles cerebral hypoxia survival time
even colder brain temperatures are required for
longer survival periods.
argues that, to explain some of the
remarkable survival stories reported, such a
mechanism of cooling lacks quantitative evidence.
On simple thermal physical principles of “methods
of mixtures”, he points out that “a (normal) person
would need to ingest an amount of ice-cold water
equivalent to 20% of body mass in order to reduce
the average body temperature to 30 ⬚C. Thus a child
weighing 20 kg would need to ingest approximately
4 litre of ice-cold water”. Such volumes of aspirated
water are not compatible with resuscitation
are they usually encountered in near-drowning
stated that it is unusual to
aspirate more than 200 ml but it is unclear how this
estimation has been determined, particularly as it
has been shown experimentally
that aspirated fresh
water is quickly absorbed and redistributed within
Gooden’s calculation, however, appears to be
based on the requirement for the entire body mass to
be cooled to 30 ⬚C. Using the same simple methods
of mixtures, the volume of ice water necessary to
cool the blood volume of a 20-kg child (approxi-
mately 1.6 litre) to 30 ⬚C would be approximately
300 ml. It is accepted that this is a gross over
simplification as the actual amount required to be
218 British Journal of Anaesthesia
aspirated to cool the brain would need to be in
excess of this because of the dynamic heat exchange
that would occur with the perivascular tissues and
perfused organs; in addition, pulmonary heat
exchange is reduced because of pulmonary shunting.
Nevertheless, given that a large proportion of the
cardiac output in a child in such circumstances (i.e.
with generalized peripheral vasoconstriction) is
diverted to the cerebral circulation, it is conceivable
that selective cooling of the brain occurs some time
before the general body mass equilibrates. Thus
aspiration of cold water may play a critical role in the
production of cerebral hypothermia and brain
survival in the submerged child or young adult of
relatively small body mass.
At first sight the volume of water required to be
aspirated would appear to be in conflict with
Modell’s conclusion that volumes in excess of 22 ml
are usually incompatible with life. This would
allow a maximum aspiration of 440 ml for a child
with a body mass of 20 kg. While this is more than
the theoretical 300 ml to cool the blood, as stated
above, larger volumes would in reality be required to
negate dynamic heat exchange with the perivascular
tissues and perfused organs. However, during the
drowning process, aspiration of water is unlikely to
occur in one single bolus but is more likely to be a
dynamic process with flushing of small volumes of
cold water in and out of the lung. Fainer, Martin and
demonstrated in dogs that rapid violent
respiratory movement continued for about 70 s after
submersion. It is proposed that with each tidal
volume of water aspirated, heat is exchanged but the
total volume retained in the lungs when respiration
ceases is significantly less than the total volume of
water which has entered and exited the lungs up to
This hypothesis is supported by a recent animal
experiment by Conn and colleagues
showed that aspiration of cold water produces
extremely rapid blood cooling; in lightly anaes-
thetized dogs, the mean carotid artery temperature
decreased by approximately 8.0 ⬚C in the first 2 min
of drowning, but only by 0.8 ⬚C in a control head-out
group of similarly shaved dogs; rectal temperature
lagged “considerably behind”. Regrettably the
experimental method did not make it possible to
estimate how much fluid had been aspirated in those
2 min. When the animals were killed after 10 min of
submersion, the volumes aspirated, estimated from
post-mortem weight changes, were considerably in
excess of the accepted lethal level of 22 ml kg
Incidentally, it is quite difficult to find the origin of
the value of 22 ml kg
quoted frequently by Modell
and others; in the original report by Modell and
the value appears to be 20 ml lb
death caused by VF rather than irreversible
The theory of water flushing in and out of the lung
rather than absorption of water as the primary cause
of the rapid cooling of blood on submersion is sup-
ported further by data from Conn and colleagues.
The marked decrease in carotid artery blood temp-
erature in dogs (average weight 9.2 kg) was 8.5 ⬚C
after 2 min in fresh water at 4 ⬚C and 7.5 ⬚C in salt
water; while over the remaining 8 min of the experi-
ment, mean carotid temperature decreased by
another 2.4–10.9 ⬚C in fresh water and by
3.1–10.6 ⬚C in salt water. Body mass increased by
1.5 kg (fresh) and 0.6 kg (salt) in the 10 min.
Although Conn and colleagues
reported that all
animals were tachypnoeic on immersion, the study
did not state when respiration ceased. In an earlier
experiment by Fainer, Martin and Ivy,
dogs, it was reported that violent respiratory
movement lasted approximately 71 s and thereafter
respiratory movement was limited to the occasional
agonal movement. Thus it would be fair to assume
that, in the experiment by Conn and colleagues,
most of the aspiration had taken place during the
2-min period showing the greatest decrease in
carotid temperature. Furthermore, if one assumes
no, or little, absorption in the salt water group in this
time, it suggests that in the equivalent fresh water
group, 7.5 ⬚C of the 8.5 ⬚C decrease in blood
temperature in the initial 2 min was caused by
flushing and 1.0 ⬚C by absorption of approximately
900 ml of fresh water.
Increasingly, hypothermia has been recognized as a
favourable prognostic indicator. However, the rate
of onset of hypothermia is clearly critical; to be of
benefit it must occur rapidly before profound
Exposure to cold ambient conditions during
resuscitative procedures, while awaiting the arrival
of the ambulance, facilitates continued cooling if
adequate insulation is not provided. Furthermore,
thermoregulation in the hypoxic individual is likely
to be severely compromised. This, combined with
high evaporate heat loss from wet clothing and
increased conductive heat loss to the ground beneath
the supine patient, cause deep body temperature to
continue to decrease. By the time the victim arrives
at hospital, body temperature, in a patient with
intact circulation, may have decreased significantly.
Unless such a decrease occurs quickly, and before
hypoxia becomes severe, little useful cerebral
protection is provided.
Thus cooling which occurs during rescue and
transport to hospital is of little benefit and may
render deep body temperature on arrival in hospital
an unreliable prognostic indicator.
Rescue and treatment
Advice on rescue lies more correctly with rescue
organizations but a few important principles are
offered for consideration:
At all times it is important to ensure the safety of
the rescuer in addition to the casualty.
! Landing a casualty should never be delayed to
enable attempts at in-water resuscitation.
! Where possible, casualties wearing life jackets
whose airway is clear of the water should be
recovered in a horizontal attitude.
Immersion, near-drowning and drowning 219
! The possibility of cervical spine trauma should be
considered in unconscious victims rescued from
shallow water and appropriate measures and
It should be presumed that all near-drowning
victims rescued from UK open waters are hypoxic,
acidotic and hypothermic.
The primary aims of treatment of the near-
drowned are listed below in order of priority:
! Effective immediate relief of hypoxia. The speed
of achieving this has the greatest influence on
! Restoration of cardiovascular stability.
! Prevention of further heat loss.
! Speedy evacuation to hospital.
Airway patency and adequate ventilation should
be achieved rapidly. There is a significant risk of
regurgitation and aspiration of gastric contents
Tracheal intubation, if
indicated, should be carried out at the earliest oppor-
High fractional inspired oxygen
should be delivered using a self-inflating bag and
reservoir; if available, a positive end-expiratory
pressure (PEEP) valve set to 5 cm H
O should also
be used. Application of cricoid pressure during
EAR, or bag and mask ventilation, reduce the risk of
regurgitation and aspiration.
The use of the Heimlich manoeuvre in near-
drowning has been considered by the Emergency
Cardiac Care Committee of the American Heart
Association and is not recommended.
insufficient evidence to support the claims that it is
effective in removing aspirated liquid. Additionally,
the manoeuvre itself may delay early intubation,
aggravate visceral or spinal injury, cause confusion
among rescuers with yet another procedure to
and increase the risk of
precipitating an undesirable cardiac arrhythmia in
the hypothermic casualty rescued from cold water.
In the hypothermic near-drowning casualty,
because of the marked cold-induced peripheral vaso-
constriction, characteristically low volume and pro-
found bradycardia, peripheral pulses are notoriously
difficult to feel. Accordingly, the carotid artery
should be palpated gently for at least 1 min.
Extreme caution must be taken not to precipitate VF
by rough handling.
VF responds poorly to defibril-
lation at core temperatures less than 28 ⬚C; at this
temperature, Purkinje tissue looses its conduction
advantage over normal ventricular muscle fibres.
Nevertheless, it may be worth attempting, especially
in children, as some successes have been
However, even when sinus rhythm is
achieved at these temperatures it tends to be
unstable and revert easily to VF. If attempts at
defibrillation at less than 28 ⬚C are not immediately
successful they should be discontinued until cardiac
temperature increases to greater than 29 ⬚C.
The casualty should be insulated against further
heat loss and transported to the nearest emergency
department as rapidly as gentle handling allows.
Any life-support measures instituted at the scene
should be continued during transportation.
authors advocate immediate transfer of all severely
hypothermic and clinically lifeless patients to
the nearest facility with cardiopulmonary bypass
Near-drowning is an emergency: even those victims
who appear normal on arrival at hospital can
An accurate, rapid initial
assessment of the victim is essential. A suggested
algorithm for subsequent management, based on
the classification by Simcock of near-drowning
is shown in figure 1.
Survival figures for victims of near-drowning are
very encouraging for those with a minimal reduction
in the Glasgow coma score (GCS) but much less
favourable for those with low scores.
Figure 1 A suggested algorithm for management of near-drowning, based on the classification by Simcock of near-
ALS:Advanced life support.
220 British Journal of Anaesthesia
which have been associated with favourable outcome
are listed in table 1, but none, either individually or
in combination, has been shown to have absolute
Therefore, full resuscitative
efforts should be attempted in all near-drowning
patients arriving in emergency departments.
Patients with no evidence of aspiration
Victims of near-drowning who appear normal in the
emergency department can deteriorate unexpect-
edly. All should be admitted to a general ward for a
period of observation, generally accepted to be a
minimum of 6 h.
Other authors, however, consider
it prudent to admit the victim for 24 h because of the
risk of late onset pulmonary and cerebral oedema.
It is recommended that the investigations listed
in table 2 are considered. Hypothermic patients who
are conscious and cooperative can be rewarmed in a
bath of warm water (40 ⬚C) until rectal temperature
has increased to 36 ⬚C. Arterial blood-gas values,
obtained while breathing air, and core temperature
must be normal before discharge. Signs of aspira-
tion, which may become manifest later, include
cough, tachypnoea, retrosternal discomfort, audible
crackles on auscultation of the chest, abnormal
arterial oxygen tension with the patient breathing air,
abnormal chest x-ray and fever.
Infection after immersion is common; usually this
is confined to the chest, although there are reports of
intracerebral abscess and septicaemia.
the benefits of prophylactic antibiotic therapy have
It is generally agreed that anti-
biotic therapy should be withheld until bacterio-
logical studies, or clinical examination, indicate
Patients should be informed of the
potential seriousness of any pyrexial illness which
develops within a few days of discharge. They should
be advised to attend either their general practitioner
or return to the A&E department in such an event.
Patients with evidence of aspiration but without
These patients may develop pulmonary oedema and
deteriorate precipitously. Accordingly, they should
be admitted to a high dependency (HDU) or inten-
sive care unit (ICU). Monitoring should include
continuous pulse oximetry and frequent arterial
blood-gas analysis. The onset of pulmonary oedema
is usually within a few hours of aspiration of water,
either fresh or salt, and can be rapidly progressive.
Continuous positive airways pressure (CPAP) sup-
port provided by a mask or nasal cannula (in infants
who are obligate, nasal breathers),
offered in the first instance. The aim is to achieve an
arterial oxygen partial pressure
(a)P of more than 8
kPa with an
less than 0.5. If this is unsuccessful
the facility for tracheal intubation and intermittent
positive pressure ventilation (IPPV) with PEEP must
be available immediately.
Patients with ventilatory compromise or hypothermia,
Patients who have clinical or investigative evidence
of inadequate ventilation must undergo tracheal
intubation and ventilatory support with PEEP.
This should be maintained at 8–10 kPa or at 10–12
kPa in those with ischaemic heart disease. Ideally,
F should be less than 0.5 and PEEP provided at
the lowest pressure consistent with minimizing
intrapulmonary shunt and achieving these aims.
All victims of near-drowning who are hypothermic
should be rewarmed in an ICU with continuous
Those who are cardiovascularly stable
may be warmed slowly but actively, at a controlled
rate of approximately 1 ⬚C h
. This can be achieved
using warmed humidified inspired gas, warmed i.v.
fluids, warming mattress/blanket or wrapping
exposed areas in dry woollen blankets. It is advisable
to consider rapid rewarming for any patient whose
core temperature is less than 28 ⬚C because of the
perilous state of the myocardium and its propensity
for ventricular arrhythmias below this temperature.
Patients who have an unstable cardiovascular system
or are in cardiopulmonary arrest should be warmed
more aggressively with consideration given to extra-
corporeal rewarming (ECR) using haemofiltration or
cardiopulmonary bypass (CPB).
Methods used to rewarm rapidly without ECR
include bladder irrigation, gastric lavage,
oesophageal heat interchanger,
thoracotomy and pleural
The easiest way to accomplish this in
most hospitals is peritoneal dialysis.
Favourable prognostic indictors for near-drowning
Children 3 yr
Duration of submersion
Time to effective basic life support
Rapid return of spontaneous cardiac output
Spontaneous cardiac output on arrival in emergency
Minimum blood pH
11.2 mmol litre
GCS: no coma on admission
Pupillary responses: present
Suggested investigations for near-drowning patients
Arterial blood gas analysis
Urea and creatinine
Total creatinine kinase
Blood and tracheal aspirate for aerobic and anaerobic culture
Rectal temperature measured at least 15 cm from the anal
Drugs screen for overdose of:
Immersion, near-drowning and drowning 221
(2 litre of potassium free) is warmed to 54 ⬚C so that
it enters the peritoneal cavity at 43.5 ⬚C, and hourly
cycles are preferred.
The advantages of ECR are rapid restoration of
normothermia up to 10.7 ⬚C h
reduction of high
blood viscosity found in hypothermia and, most
importantly, when bypass is used, there is reinstitu-
tion of perfusion regardless of cardiac rhythm
avoidance of “rewarming shock”. An additional
benefit is the ability to rapidly off-load ultrafiltrate in
the presence of overwhelming pulmonary oedema;
Norberg and colleagues removed 2300 ml in 1 h to
facilitate withdrawal from CPB.
Those in cardiopulmonary arrest
Full resuscitation must be attempted in accordance
with the Resuscitation Council (UK) guidelines and
an attempt made to rewarm the casualty to 33 ⬚C.
This may cause manpower problems when pro-
longed external chest compressions (ECC) are
required. Successful manual massage maintained for
a period of 4.5 h has been demonstrated,
Ireland and colleagues reported success after 5.5 h
using a “Thumper” cardiopulmonary resuscitator
system (Michigan Instruments Inc., Grand Rapids,
MI, USA) mechanical device.
The advantage of a
mechanical device is to ensure consistent effective
compression during rewarming thus maintaining an
adequate circulation to meet metabolic demands
and facilitate regular emptying of the heart during
Jones and Swan also recommend the use of
such devices in prolonged resuscitation.
Rewarming by institution of emergency CPB has
been used with dramatic effect in severely hypo-
thermic casualties who were in full arrest,
11 43 47 49 98
many after submersion in cold water.
50 60 72 99
most notable of these was the neurologically intact
survival of a 2.5-yr-old girl who was trapped under
ice for 66 min.
The majority of reports describe partial bypass
from the femoral artery to femoral vein.
11 43 49 98
Norberg and colleagues described conversion to full
bypass to achieve faster flow rates and decompres-
sion of a dilating asystolic heart in a 7-yr-old boy
with a successful outcome.
Letsou and co-workers
suggested that full bypass using median sternotomy
is the method of choice.
There is also a report of
successful resuscitation in one of two hypothermic
adult casualties using a modified dilatational/
Seldinger technique to apply percutaneous CPB via
the femoral veins, with a portable pump and
CPB provides fast
rewarming of the core, maintains tissue perfusion
and oxygenation while assisting the cold and
inefficient heart to regain normal electromechanical
The advocates of median sternotomy claim that
exposure and cannulation of the aorta and right
atrium is faster than dissection and cannulation of
the femoral vessels in the pulseless patient, and that
flow rates are faster. There may also be advantage in
having access to the heart for massage, defibrillation
and decompressive venting of a cold and dilating
A survey of members of the American
Association of Thoracic Surgeons and the Southern
Thoracic Surgical Association in the early 1980s
produced evidence of 174 cases of hypothermia
rewarmed with CPB with a survival rate of 45%.
review of the literature in 1994
analysed data from
68 hypothermic patients undergoing CPB. In
reports describing attempted resuscitation of two or
more cases the survival rate was 50%; if case reports
and singletons were included the rate improved to
60%. VF was associated with greater survival than
asystole, but the difference (73% vs 52%) was not
significant. Walpoth and colleagues also noted that
documented VF had a favourable outcome with
CPB in hypothermia.
Eighty percent of survivors
returned to their previous level of function. The
recommendations of that review were that CPB
resuscitation is advised for hypothermic patients in
arrest and for those with core temperatures less than
25 ⬚C, irrespective of rhythm; patients in a stable
rhythm of 25–28 ⬚C can be treated with CPB or con-
ventional warming methods.
The maxim “nobody
is dead until warm and dead” should always be
borne in mind.
Cessation of resuscitative efforts
Hyperkalaemia in severely hypothermic patients is
usually indicative of asphyxial cardiac arrest before
significant cooling occurred.
A potassium concen-
tration in excess of 10 mmol litre
is not compatible
with successful resuscitation
and is advocated for
use in the triage of avalanche victims.
Hypoxic encephalopathy is probably the most
debilitating of the consequences of near-drowning.
The majority of flaccid, comatose victims admitted
to the ICU suffer either severe cerebral injury or
cerebral death. Early prediction of outcome requires
accurate determination of the nature and extent of
neurological injury; it is also fundamental for patient
management. In children, there are four distinct
outcomes: full recovery, neurologically impaired
survival, persistent vegetative state and death.
Cerebral resuscitative efforts to reverse the changes
of hypoxic damage and reperfusion injury have rarely
Poor outcomes are especially
likely when ICP 920 and CPP :60 mm Hg, despite
One USA source
cited by Orlowski
that one-third of all near-drowning survivors were
moderately to severely brain damaged. The progno-
sis for children in the UK appears to be similar: of
188 hospital resuscitations of near-drowning
patients reported by Kemp and Sibert,
fixed dilated pupils on admission to hospital; of these
13 died, 10 had severe residual neurological defects
(spastic quadriplegia) and 10 recovered fully. A
recent report from Seattle
found an unfavourable
outcome in patients who were comatose on admis-
sion: of 72 admissions there were 38 deaths, 14 in a
222 British Journal of Anaesthesia
vegetative state, three with dependent neurological
impairment, six with mild neurological impairment
and 11 discharged normal. The Australian experi-
ence was similar: 31 admissions with cardiac arrest
of whom 17 died and six survived with severe spastic
This could be related to the warmer
waters found in the Brisbane area. This view is
supported by the Canadian experience
prolonged in-hospital resuscitation and aggressive
treatment of near-drowning victims who initially had
absent vital signs and were not hypothermic
(933 ⬚C), resulted either in eventual death or an
increase in the number of survivors with a persistent
A prognosis score for paediatric near-drowning
victims has been derived by Orlowski.
It is based
on five indicators: age younger than 3 yr, maximum
submersion estimated longer than 5 min, post-
rescue delay in resuscitation for greater than 10 min,
coma at time of admission to hospital and initial pH
less than or equal to 7.1.
Neurological classification of 101 paediatric
drowning and near-drowning victims based on the
level of consciousness, as assessed by GCS, has been
used for prognostic purposes.
GCS:5 predicted a
high-risk group of patients with 80% mortality or
permanent neurological sequelae. In general, 2–6 h
after the accident, patients who remain decorticate,
decerebrate or flaccid fair poorly; patients who are
improving at this time, but remain unresponsive,
have a 50% chance of doing well, while those who
are showing definite signs of improvement (alert,
stuporous or obtunded but responding to painful
stimuli) generally do well and most have a normal or
near normal outcome.
A recent article by Kreis and
suggested that the most accurate
method of predicting neurological outcome is by
magnetic resonance spectroscopy (MRS): their
spectroscopic prognosis index distinguished between
good (n:5) and poor (n:11) outcome, with one
false negative after a borderline MRS result, and no
false positive results (100% specificity).
It is generally agreed that hypothermia is neuro-
logically protective, especially in children, although
Kruus and co-workers concluded that a core
temperature of less than 30 ⬚C is indicative of pro-
longed immersion and therefore of more severe brain
In reports where rapid reversal of the
hypoxic state was achieved, in combination with
ICU support, often with ECR and occasionally with
control of intracranial pressure (ICP), there was
complete or near-complete return of neurological
function with time.
11 85 86 102
Many studies assessing predictive factors have
analysed data from near-drownings in warmer
waters than are usually found in the UK.
Therefore, it may be inappropriate to assess
immersion victims using the same criteria, as the
presence of hypothermia may offer some cerebral
In addition to the complications already described
(pulmonary oedema, infection, hypothermic cardiac
arrhythmia, hypoxic encephalopathy and cerebral
oedema), disseminated intravascular coagulation
(DIC) has been described in near-drowning,
Rhabdomyolysis complicated by acute renal
failure has been reported in two normothermic
casualties of near-drowning in cold (14 ⬚C and
17 ⬚C) surf.
One victim who was retrieved in cardio-
pulmonary arrest and was alert, lucid and apparently
normal 1 h later, had a plasma creatinine concentra-
tion of 2106 mol litre
3 h after immersion. He
required support with peritoneal dialysis for 11 days
and forced saline diuresis to avert oliguria. The
second patient discharged 2 h after the event, re-
presented 24 h later complaining of weakness,
myalgia and dark urine. His serum creatinine
concentration was 359 mol litre
, and forced
saline/frusemide diuresis avoided the need for dialy-
sis. Excessive or sustained muscle exercise and heat
stress are well recognized causes of rhabdomyolysis;
cold stress is a significantly less common trigger.
However, in view of the potential consequence of
acute renal failure, it would seem prudent to
measure total creatinine kinase activity in addition to
serum electrolytes and simple urinalysis in survivors
Peripheral neurological sequelae may be manifes-
tations of traumatic or even “non-freezing cold
injury” to peripheral nerves rather than hypoxic
damage from near-drowning. One should also be
mindful of the near-drowning patient having a pre-
disposing injury or medical condition, while the
possibility of there being a drug-related condition
should also be borne in mind.
Other treatment modalities
Artificial surfactant (Exosurf
) has been used success-
fully in the management of a 9-yr-old victim of near-
drowning with a rectal temperature of 30.3 ⬚C who
was acidotic (pH 6.95).
Despite circulatory and
ventilatory support (
F 1, minute volume 230 ml
, PEEP 5 cm H
O, peak ventilator airway pres-
sure 40 cm H
O) acidosis persisted and there was
severe and progressive respiratory distress (pH 6.87,
aP 7.6 kPa,
aP 12.4 kPa, base excess 917.5
). Four hours after admission artificial
surfactant was instilled slowly via the tracheal tube.
Within 5 min there was an improvement in gas
aP 11.5 kPa,
aP 9.5 kPa) and a
decrease in peak pressure to 30 cm H
O, with no
change in ventilator settings. His trachea was
extubated 3 days later and he was discharged
Animal models of near-drowning have failed to
demonstrate any improvement in pulmonary
function after the use of exogenous surfactant,
4 7 76
although damage to the alveolar capillary
membranes is reduced. This disappointing finding
may be a result of using volumes of surfactant which
were too large and instilled too soon after injury,
causing documented severe hypoxia (i.e. a second
major insult). The use of smaller volumes, instilled
or nebulized at different times, with additional
PEEP, have not been reported.
Immersion, near-drowning and drowning 223
A study on the effect of corticosteroid therapy on
survival and blood-gas exchange in dogs after
aspiration of fresh water drowning demonstrated no
Their value in the management
of pulmonary injury has not been proved
probably should not be used.
The rapid onset of cerebral hypothermia would
appear to be the essential mechanism by which some
submerged victims, in particular children, appear to
be able to withstand protracted periods of asphyxia
without suffering irreversible neurological damage.
For this protective mechanism to be effective, brain
cooling must be established rapidly, before severe
hypoxia occurs. Such rates of cooling are unlikely to
be achieved through surface cooling alone, or in
combination with the diving reflex. It is postulated
that rapid selective brain cooling is achieved through
pulmonary heat exchange by repeated flushing of the
lungs with cold water during the drowning process
before cardiorespiratory arrest supervenes.
Successful treatment depends on speedy recovery
and the immediate institution of CPR if indicated.
All near-drowning victims, even those who appear to
be ostensibly clinically well, should be transferred to
hospital for pulmonary screening. Although some
successful resuscitations of victims in cardiac arrest
who are flaccid with fixed dilated pupils have been
reported, the great majority would appear to have a
poor prognosis. The prognosis is better for those
with profound hypothermia but rewarming may
be difficult in the absence of a good circulation.
Accordingly, consideration should be given to trans-
ferring all profoundly hypothermic patients who are
in cardiac arrest to a centre where cardiopulmonary
bypass facilities are available, sooner rather than
1. Adams RD, Victor M. Hypoxic hypotensive encephalo-
pathy. In: Adams RD, Victor M, eds. Principles of Neurology.
New York: McGraw-Hill Book Co, 1977; 732–734.
2. Agar JWM. Rhabdomyloysis and acute renal failure after
near-drowning in cold salt water. Medical Journal of
Australia 1994; 161: 686–687.
3. Alexander L. The treatment of shock from prolonged
exposure to cold, especially in water. Combined Intelligence
Objectives Sub-committee APO 413 C105, Item No. 24.
London: HMSO, 1945.
4. Anker AL, Santora T, Spivey W. Artificial surfactant
administration in an animal model of near drowning.
Academic Emergency Medicine 1995; 2: 204–210.
5. Arnet EL, Watts DT. Catecholamine excretion in men
exposed to cold. Journal of Applied Physiology 1960; 15:
6. Ashwal S, Schneider S, Tomasi L, Thompson J. Prognostic
implications of hyperglycemia and reduced cerebral blood
flow in childhood near-drowning. Neurology 1990; 40:
7. Bambang OS, Reijngoud DJ, Ennema JJ. Surfactant
replacement therapy in surfactant-deficient rabbits: early
effective lung function and biochemical aspects. Lung 1988;
8. Bell TS, Ellenberg L, McComb JG. Neuropsychological
outcome in severe pediatric near-drowning. Neurosurgery
1985; 17: 604–608.
9. Bierens JJ, van der Velde EA, van Berkel M, van Zanten JJ.
Submersion in the Netherlands: prognostic indicators and
results of resuscitation. Annals of Emergency Medicine 1990;
10. Biggart MJ, Bohn DJ. Effect of hypothermia and cardiac
arrest on outcome of near-drowning accidents in children.
Journal of Pediatrics 1990; 117: 179–183.
11. Bolte RG, Black PG, Bowers RS, Kent-Thorne J, Correli
HM. The use of extracorporeal rewarming in a child sub-
merged for 66 minutes. Journal of the American Medical
Association 1988; 260: 377–379.
12. Brugger H, Durrer B, Adler-Kastner L. On-site triage of
avalanche victims with asystole by the emergency doctor.
Resuscitation 1996; 31: 11–16.
13. Calderwood HW, Modell JH, Ruiz BC. The ineffectiveness
of steroid therapy for treatment of fresh-water near-drown-
ing. Anesthesiology 1975; 43: 642–650.
14. Cavalier S, Drownings. National Pool and Spa Safety
Conference. Washington, DC: US Consumer Product Safety
Commission, 1985; 9–12.
15. Conn AW, Barker GA, Edmonds JF, Bohn MB.
Submersion hypothermia and near-drowning. In: Pozos RS,
Wittmers LE, eds. The Nature and Treatment of Hypothermia.
Minneapolis, MN: University of Minnesota Press, 1983;
16. Conn AW, Miyassaka K, Katayama M, Fujita M, Orima H,
Barker G, Bohn D. A canine study of cold water drowning
in fresh versus salt water. Critical Care Medicine 1995; 23:
17. Cooper KE, Ross DN. Hypothermia in Surgical Practice.
London: Cassell, 1960.
18. Daly M deB, Angell-James JE, Elsner R. Role of carotid-
body chemoreceptors and their reflex interactions in brady-
cardia and cardiac arrest. Lancet 1979; 1: 764–767.
19. Dean JM, Kaufman ND. Prognostic indicators in paediatric
near-drowning: The Glasgow Coma Scale. Critical Care
Medicine 1981; 9: 536–539.
20. Derbyshire DR, Clark RG. Cerebral recovery after pro-
longed global brain ischaemia. Lancet 1980; 2: 637.
21. DeVillota ED, Barat G, Peral P, Juffe A, de Miguel JMF,
Avello F. Recovery from profound hypothermia with
cardiac arrest after immersion. British Medical Journal 1973;
22. Edwards ND, Timmins AC, Randalls B, Morgan GAR,
Simcock AD. Survival in adults after cardiac arrest due to
drowning. Intensive Care Medicine 1990; 16: 336–337.
23. Elsner R, Gooden BA. Diving and asphyxia—a comparative
study of animals and man. Cambridge: Cambridge
University Press, 1983.
24. Epstein M. Renal effects of head-out water immersion in
man: implications for understanding of volume haemostasis.
Physiology Reviews 1978; 58: 529–581.
25. Epstein M, Duncan DC, Fishman LM. Characterisation of
the natriuresis caused in normal man by immersion in
water. Clinical Science 1972; 43: 275–287.
26. Fainer DC, Martin CG, Ivy AC. Resuscitation of dogs from
fresh water drowning. Journal of Applied Physiology 1951; 3:
27. Fuller RH. Drowning and the post-immersion syndrome: a
clinicopathological study. Military Medicine 1964; 128:
28. Golden FStC. Recognition and treatment of immersion
hypothermia. Proceedings of the Royal Society of Medicine
1973; 66: 16–19.
29. Golden FStC. Problems of immersion. British Journal of
Hospital Medicine 1980; 23: 371–383.
30. Golden FStC, Hampton IFG, Hervey GR, Knibbs AV.
Shivering intensity in humans during immersion in cold
water. Journal of Physiology (London) 1979; 290: 48P.
31. Golden FStC, Hardcastle PT. Swimming failure in cold
water. Journal of Physiology (London) 1982; 330: 60–61P.
32. Golden FStC, Hervey GR, Tipton MJ. Circum-rescue
collapse: collapse, sometimes fatal, associated with rescue of
immersion victims. Journal of the Royal Naval Medical
Society 1991; 77: 139–149.
33. Gooden BA. Why some people do not drown; hypothermia
versus the diving response. Medical Journal of Australia
1992; 157: 629–632.
224 British Journal of Anaesthesia
34. Graf WD, Cummings PC, Quan L, Brutacao D. Predicting
outcome in pediatric submersion victims. Annals of
Emergency Medicine 1995; 2446: 312–319.
35. Gregory RT, Patton JF. Treatment after exposure to cold.
Lancet 1972; 1: 377.
36. Harries MG. Drowning in man. Critical Care Medicine 1981;
37. Harries MG. Near drowning. In: Colquhoun MC, Handley
AJ, Evans T eds. ABC of Resuscitation. London: BMJ, 1995;
38. Harries MG, Golden FS, Fowler M. Ventricular fibrillation
as a complication of salt-water immersion. British Medical
Journal 1981; 283: 347–348.
39. Hauty MG, Barry C, Esrig MD, Jonathan G, Hill MD,
Long WB. Prognostic factors in severe accidental hypo-
thermia: Experience from the Mt Hood tragedy. Journal of
Trauma 1987; 10: 1107–1112.
40. Hayward JS, Erickson JD. Physiological responses and
survival time predictions for humans in ice-water. Aviation
Space and Environmental Physiology 1984; 55: 206–212.
41. Hayward JS, Hay C, Matthews BR, Overweel CH, Radford
DD. Temperature effect on the human dive response in
relation to cold water near-drowning. Journal of Applied
Physiology 1984; 56: 202–206.
42. Hernandez MJ. Cerebral circulation during hypothermia.
In: Pozos RS, Wittmers LE, eds. The Nature and Treatment
of Hypothermia. Minneapolis, MN: University of Minnesota
Press, 1983; 152–164.
43. Hill JG, Bruhn PS, Cohen SE, Gallagher MW, Manart F,
Moore CA, Seifert PE, Askari P, Banchieri C. Emergent
applications of cardiopulmonary support: a multi-
institutional experience. Annals of Thoracic Surgery 1992;
44. Home Office. Report on the working Party on water safety.
London: HMSO, 1977.
45. Hunt PK. Effect and treatment of the “diving reflex”.
Canadian Medical Association Journal 1974; 111:
46. Hunter AR. Neurosurgical Anaesthesia, 2nd Edn. Oxford:
Blackwell Scientific Publications, 1975.
47. Ireland AJ, Pathi VL, Crawford R, Colquhoun IW. Back
from the dead: extracorporeal rewarming of severe
accidental hypothermia victims in accident and emergency.
Journal of Accident & Emergency Medicine 1997 (in press).
48. Johnson DG, Hayward JS, Jacobs TP, Collis ML,
Eckersman JD, Williams RH. Plasma norepinephrine
responses of man in cold water. Journal of Applied Physiology
1977; 43: 216–220.
49. Jones AI, Swan IJ. Prolonged resuscitation in accidental
hypothermia: use of mechanical cardio-pulmonary resusci-
tation and partial cardio-pulmonary bypass. European
Journal of Emergency Medicine 1994; 1: 34–36.
50. Kangas E, Niemela H, Kojo N. Treatment of hypothermic
circulatory arrest with thoracotomy and pleural lavage.
Annales Chirurgiae et Gynaecologiae 1994; 83: 258–260.
51. Keatinge WR, McIlroy MB, Goldfien A. Cardiovascular
responses to ice-cold showers. Journal of Applied Physiology
1964; 19: 1145–1150.
52. Keatinge WR, Nadel JA. Immediate respiratory response to
sudden cooling of the skin. Journal of Applied Physiology
1965; 20: 65–69.
53. Kemp AM, Sibert JR. Outcome for children who nearly
drown: a British Isles study. British Medical Journal 1991;
54. King RB, Webster IW. A case of recovery from drowning
and prolonged anoxia. Medical Journal of Australia 1964; 1:
55. Knight DR, Horvarth SM. Urinary responses to cold tem-
peratures during water immersion. American Journal of
Physiology 1985; 248: R560–566.
56. Kreis R, Arcinue E, Ernst T, Shonk TK, Flores R, Ross BD.
Hypoxic encephalopathy after near-drowning studied by
quantitative 1H-magnetic resonance spectroscopy: meta-
bolic changes and their prognostic value. Journal of Clinical
Investigation 1996; 97: 1142–1154.
57. Kristensen G, Gravesen M, Bemveniste D, Jordening H. An
oesophageal thermal tube for rewarming in hypothermia.
Acta Anaesthesiologica Scandinavica 1985; 29: 846–848.
58. Kruus S, Bergstrom L, Suutarinen T, Hyvonen R.
Prognosis of near-drowned children. Acta Paediatrica
Scandinavica 1979; 68: 315–322.
59. Kvittingen TD, Naess A. Recovery from drowning in fresh
water. British Medical Journal 1963; 1: 1315–1317.
60. Letsou GV, Kopf GS, Elefteriades JA, Carter JE, Baldwin
JC, Hammond GL. Is cardiopulmonary bypass effective for
treatment of hypothermic arrest due to drowning or
exposure? Archives of Surgery 1992; 127: 525–528.
61. Lollgen H, von Neiding G, Krekelar H, Smidt V,
Kroppenhagen K, Frank H. Respiratory gas exchange and
lung perfusion in man during head-out water immersion.
Undersea Biomedical Research 1976; 3: 49–56.
62. McBrien M, Katumba JJ, Mukhtar AI. Artificial surfactant
in the treatment of near drowning. Lancet 1993; 342:
63. Mair P, Kornberger E, Furtwaengler W, Balogh D,
Antretter H. Prognostic markers in patients with severe
accidental hypothermia and circulatory arrest. Resuscitation
1994; 27: 47–54.
64. Modell JH. Pathophysiologgy and Treatment of Drowning.
Springfield, Illinois: Charles C Thomas, 1971.
65. Modell JH. Drown versus near-drown: a discussion of
definitions. Critical Care Medicine 1989; 9: 351–352.
66. Modell JH. Drowning. New England Journal of Medicine
1993; 328: 253–256.
67. Modell JH, Davis JH. Electrolyte changes in human drown-
ing victims. Anesthesiology 1969; 30: 414–420.
68. Modell JH, Graves SA, Ketover A. Clinical course of 91
consecutive near-drowning victims. Chest 1976; 70:
69. Modell JH, Moya F. Effects of volume of aspirated fluid
during chlorinated fresh water drowning. Anesthesiology
1966; 27: 662–672.
70. Mohri H, Dillard DH, Crawford EW, Martin WE,
Merendino KA. Method of surface induced deep hypo-
thermia for open-heart surgery in infants. Journal of Thoracic
Cardiovascular Surgery 1969; 58: 262–270.
71. Nemiroff MJ, Saltz GR, Weg JG. Survival after cold-water
near-drowning: the protective effect of the diving reflex.
American Review of Respiratory Diseases 1977; 115: 145.
72. Norberg WJ, Agnew RF, Brunsvold R, Sivanna P, Browdie
DA, Fisher D. Successful resuscitation of a cold water sub-
mersion victim with the use of cardiopulmonary bypass.
Critical Care Medicine 1992; 20: 1355–1357.
73. Orlowski JP. Prognostic factors in pediatric cases of drown-
ing and near drowning. Journal of the American College of
Emergency Physicians 1979; 8: 176–179.
74. Orlowski JP. Drowning, near-drowning and ice-water sub-
mersions. Pediatric Clinics of North America 1987; 34: 75–92.
75. Orlowski JP. Drowning, near-downing, and ice-water
drowning. Journal of the American Medical Association 1988;
76. Orlowski JP, Abulleil MM, Phillips JM. The hemodynamic
and cardiovascular effects of near drowning in hypotonic,
isotonic, or hypertonic solutions. Annals of Emergency
Medicine 1989; 18: 1044–1049.
77. Pearn JH. Secondary drowning in children. British Medical
Journal 1980; 281: 1103–1105.
78. Pearn JH. The management of near-drowning. British
Medical Journal 1985; 291: 1447–1452.
79. Phillips SJ, Zeff RH, Kongtahworn C, Skinner JR, Toon
RS, Grignon A, Kennerly RM, Wickemeyer W, Iannone
LA. Percutaneous cardiopulmonary bypass: application and
indication for use. Annals of Thoracic Surgery 1989; 47:
80. Ramey CA, Ramey DN, Hayward JS. Dive response of
children in relation to cold-water near-drowning. Journal of
Applied Physiology 1987; 63: 665–668.
81. Report of the Committee on Accidental Hypothermia.
London: Royal College of Physicians of London, 1966.
82. Rivers JF, Orr G, Lee HA. Drowning: its clinical sequelae
and management. British Medical Journal 1970; 2: 157–161.
83. Rosen P, Stoto M, Harley J. The use of the Heimlich
manoeuvre in near drowning: Institute of Medicine report.
Journal of Emergency Medicine 1995; 13: 397–405.
84. Schaller MD, Fischer AP, Perret CH. Hyperkalemia: a
prognostic factor during severe hypothermia. Journal of the
American Medical Association 1990; 264: 1842–1845.
Immersion, near-drowning and drowning 225
85. Sekar S, MacDonnell KF, Namsirikul P, Herman RS.
Survival after prolonged submersion in cold water without
neurological sequelae. Archives of Internal Medicine 1980;
86. Siebke H, Rod T, Breivik H, Lind B. Survival after 40
minutes submersion without cerebral sequelae. Lancet
1975; i: 1275–1277.
87. Simcock AD. Treatment of near drowning—a review of 130
cases. Anaesthesia 1986; 41: 643–648.
88. Simcock AD. Current aspects of near drowning. Applied
Cardiopulmonary Pathophysiology 1991; 3: 327–332.
89. Stone HH, Donnelly C, Frosbese AS. The effect of lowered
body temperature on cerebral hemodynamics and
metabolism of man. Surgery, Gynecology and Obstetrics 1956;
90. Stoneham MD, Squires SJ. Prolonged resuscitation in acute
deep hypothermia. Anaesthesia 1992; 47: 784–788.
91. Tipton MJ. The initial responses to cold water immersion in
man. Clinical Sciences 1989; 77: 581–588.
92. Tipton MJ. The effect of clothing on “diving bradycardia”
in man during submersion in cold water. European Journal of
Applied Physiology 1989; 59: 360–364.
93. Tipton MJ, Golden FStC. The influence of regional insula-
tion on the initial responses to cold immersion. Aviation
Space and Environmental Physiology 1987; 58: 1192–1196.
94. Tipton MJ, Golden FStC. Immersion in cold water: effects
on performance and safety. In: Harries M, Williams C,
Stanish WD, Micheli LJ, eds. Oxford Textbook of Sports
Medicine. New York: Oxford University Press, 1994;
95. Tipton MJ, Kelleher PC, Golden FStC. Supraventricular
arrhythmias following breath-hold submersions in cold
water. Undersea and Hyperbaric Medicine 1994; 21:
96. Tipton MJ, Stubbs DA, Elliott DS. Human initial responses
to immersion in water at 3 temperatures and following
hyperventilation. Journal of Applied Physiology 1991; 70:
97. Vretenar DF, Urschel JD, Parrott JCW, Unruh HW.
Cardiopulmonary bypass resuscitation for accidental hypo-
thermia. Annals of Thoracic Surgery 1994; 58: 895–898.
98. Walpoth BH, Locher T, Leupi F, Schüpbach P,
Mühlemann, Althaus U. Accidental deep hypothermia with
cardiopulmonary arrest: extracorporeal blood rewarming in
11 patients. European Journal of Cardio-thoracic Surgery
1990; 4: 390–393.
99. Waters DJ, Belz M, Lawse D, Ulstad D. Portable cardio-
pulmonary bypass: resuscitation from prolonged ice-water
submersion and asystole. Annals of Thoracic Surgery 1994;
100. Waugh JH, O’Callaghan MJ, Pitt WR. Prognostic factors
and long-term outcomes for children who have nearly
drowned. Medical Journal of Australia 1994; 161:
101. Whitelaw WA, McBride B, Armar J, Corbet K. Respiratory
neuromuscular output during breath-holding. Journal of
Applied Physiology 1981; 50: 435–443.
102. Young RSK, Zalnaratis EL, Dooling EC. Neurological out-
come in cold water drowning Journal of the American Medical
Association 1980; 244: 1233–1235.