Malaysian J Path01 1992; 14(2):77 - 83

REVIEW
Drowning
R Sarvesvaran DMJ (Lond), MD (For. Med)

Department of Pathology, University of Malaya, Kuala Lumpur

Abstract
A body recovered from the water does not necessarily imply that death was due to drowning. The
diagnosis of drowning is discussed together with the significance of the "diatom" and biochemical
tests.

Key words: Cadaveric spasm, "Champignon de mousse", drowning, diatoms, fresh water, sea water.

INTRODUCTION and resuscitation procedures continued for some

time before life is pronounced extinct.
Drowning is the result of immersion, partial or
complete, in a fluid medium. The medium is
GENERAL OBSERVATION
usually water but James' reported a case of
drowning in a vat of beer. The death rate of A. External features
drowning has been estimated in the U.S.A. as
Hypostusis
2.5 per 100,000 p ~ p u l a t i o n . ~
With the increase in popularity of boating Post mortem hypostasis may be confined to the
and water sports, a larger population is at risk. head, neck and front of the chest. This distribu-
Although complete immersion is usual, drowning tion is because a fully submerged body tends to
may occur when the nostrils and mouth only are float with the face downwards and the buttocks
covered by fluid. In cases where immersion is upwards. Movement of the body in water will
"partial", a predisposing factor must be consid- influence the development of hypotasis and thus
ered (e.g. loss of consciousness due to head may not be obvious in these circumstances. The
injury, an epileptiform attack, myocardial hypostasis may have a pink tinge suggesting
infarction or alcoholic intoxication). carbon monoxide poisoning but distinguishable
from it by its distribution and the absence of
CLASSIFICATION OF DROWNING carbon monoxide in the blood. The colour is
due to exposure and oxygenation of the depen-
This state can be morphologically classified into
dent blood and its distribution is determined by
(a) dry (atypical) drowning, (b) wet drowning,
the position of the body as it floats in the water.
and (c) secondary drowning."
In "wet" drowning the classical features of
drowning are evident, whilst in "dry" or atypical
Cutis anserina
drowning, there are little or none of the features Goose-flesh (cutis anserina) may be present. It
associated with typical drowning. In such cases, can occur between somatic and molecular death
death is attributed to either vagal stimulation and is found in bodies recovered from water and
(cardiac inhibition) or laryngeal spasm due to also in circumstances other than drowning and
submersion. In secondary drowning, the victim hence is of no diagnostic significance.
is recovered from water alive but succumbs
from about an hour to several weeks later. This Washer-woman hands
type is usually associated with metabolic acido-
sis, pulmonary oedema or chemical pneumonitis. S o d d e n , wrinkled hands and f e e t -
In cases where the body has been recovered "Washerwomen's hands" (Fig. 1) occur as a
from water, especially cold water, the possibil- result of submersion in water, irrespective of
ity of suspended animation must be considered whether the body was immersed before or after
death. The earliest skin changes - whitening and

Address for correspondence andreprrnt requests: Dr. R Sawesvaran, Department of Pathology, Faculty of Medicine, Univers~tyof Malaya. 59100 Kuala
Lumpur, Malays~a.
Malaysian J Pathol December 1992

FIG. 1 : Hand showing "washer-woman" effect.

wrinkling - appear on the back of the hands in

about 5-8 hours in summer, and 3-5 days in
winter." As the duration of immersion in water
increases, by the end of the first week early
signs of separation of the skin of the pulps of
fingers and toes occur with complete separation
including the nails occurring in about 3-4 weeks.
These changes may be delayed during the winter
months and hastened during the summer period.
These changes in the feet will be delayed if the
deceased wears shoes.

"Champignon de mousse"
The presence of foam at the mouth and nostrils
is an important sign of drowning. particularly if
the body is freshly recovered from water and has
not undergone putrefaction and other causes of
pulmonary oedema can be excluded (Fig. 2).
Other causes of pulmonary oedema are epileptic
fits, acute left ventricular failure and toxic fumes.
This foam is "fine froth", white or pink in
colour and, when wiped away, more may appear.
Even if not evident externally, it can be seen in
the upper air passages. The production of te-
nacious foam in drowning is essentially a vital
phenomenon. The entry of fluid into the air
passages provokes them to produce mucus. This
substance when mixed with water and possibly
surfactant from the lungs is readily whipped into FIG. 2: "Champignon de mousse" in a recently
a tenacious foam by the violent respiratory ef- recovered body from a river.
forts of the victim." Froth may also appear around
the mouth and nostrils in advanced states of
putrefaction. Under these circumstances the

78
 DROWNING

foam is of a coarse appearance and blood stained. cardiac inhibition.

The interpretation of such "froth" must be The mechanism of production of emphysema
guarded and no firm conclusions can be drawn aquosum is related to the production of fine
from it. foam (already discussed). When sufficient foam
has been produced, it acts like a check valve
Cadaveric spasm when powerful inspiratory efforts carry the air
past the obstructions bur the expiratory efforts
This uncommon event, if present, with weeds or
are insufficient to expel the air, water and foam.
other material from the river bed or bank firmly In the heart, the right ventricle and atria may
grasped in the hand, provides valuable evidence be dilated and the great veins distended with
of life at the time of submersion. fluid blood. The aorta (in a "fresh" body) may
show haemolytic staining of the intima.
Nail scrapings
This is a practice not routinely carried out, but if C. Dry (atypical drowning)
sand, mud or other material similar to that present The incidence of this type of drowning is re-
at the 'scene' were found, this would suggest ported as 10- 15% of cases.' When the classical
contact with it either before or after death. features of "wet" drowning are not present and
other causes of death have been excluded viz
Injuries natural or deaths due to violence - death could
Injuries may be the result of an incident prior to be attributed to dry drowning, the mechanism
entry into water or during or after submersion. being either due to vagal stimulation resulting in
Injuries sustained after submersion can be the cardiac inhibition or laryngeal spasm.
result of being attacked by fish etc, or the result
of contact with boats (propeller), etc. The Vagal stimlrla[ion
presence or absence of vital reaction would be The sudden inrush of water into the nasophar-
of assistance in differentiating these injuries but ynx or larynx may be responsible for this phe-
it must be remembered that head injuries sus- nomenon. Gardinel-8 and Spilsbury9 have re-
tained after death may bleed and resemble ante- ported such deaths. Surprise or the extreme
mortem i n j ~ r i e s . ~ coldness of the water may play an important role
in the causation of this reflex effect.
B. Internal features
The internal findings of drowning are obscured Laryngeal spusm
or abolished if putrefactive changes are advanced. It is likely that in deaths due to drowning,
The brain, liver, kidney and pancreas show varying degrees of laryngeal spasm occurs. In
no specific features. such instances, the spasm is a transient or in-
Most of the significant findings are in the termittent factor. In those circumstances when
respiratory system with some non-specific death is attributable solely to laryngeal spasm
changes in the cardio-vascular system. The air the features of mechanical asphyxia will be
passages may contain fine foam (already men- evident. The lungs are not waterlogged with
tioned) and a variable quantity of water. The
little or no fluid in the air passages. Since dry
larynx may be ~ o n g e s t e d . ~The pleurae may drowning is a rare mode of death from submer-
show haemorrhages but not the pinpoint petechial
sion, care must be taken to exclude mechanical
haemorrhages of the asphyxial type, which are causes of asphyxia when asphyxial signs are
infrequent4. These haemorrhages may be due to prominent. Modelllo suggested that 10% of
tearing of the interalveolar partitions 6 drowning victims die of asphyxia due to
The lungs are bulky and ballooned and tend
laryngospasm.
to overlap the pericardium. Rib markings may
also be present. The 1ur.g weight is increased TIME TAKEN TO DROWN
and on sectioning, war-ry fluid exudes, which
can be blood stained. The state of the lungs in The time taken to drown varies widely depending
drowning has been described as "emphysema on a number of factors: the nervous state,
aquosum". The degree of emphysema aquosum physique, reaction to submersion, ability/inability
will depend on the time taken to drown, being to swim, temperature of the water, the volume of
absent if unconsciousness or death resulted from water inhaled and the consistency of the water
(saltlfresh); death being almost immediate when
Maluysian .l Puthol December. 1992

due to cardiac inhibition, and more rapid in holding time runs the risk of losing conscious-
fresh water than in sea water. It is also rapid in ness and drowning follows quickly.
those who are unable to swim, with poor phy- At the end of such effort the arterial PO, is low
sique and when submersion is unexpected. but his deliberately reduced pCO, delays the
onset of respiratory efforts. Peripheral pooling
MECHANISM OF DROWNING of blood occurs when exertion ceases and car-
diac output falls resulting in cerebral hypoxia,
(I) Fresh watet
this accounting for the loss of consciousness and
The inhalation of water results in its absorption death by drowning.
by the circulating blood. This can occur within Water-skiing accidents can also lead to death
a few minutes following total submersion with by drowning, particularly if a severe head, or
coincident haemolysis. The heart muscle sustains skeletal or organ injury is sustained."
"serious biochemical insult,"" there being an
increase in plasma potassium with sodium loss. DIATOMS AND DROWNING
The haemodilution also overloads the circulation.
Diatoms are microscopic unicellular or colonial
This, along with myocardial anoxia, is respon-
algae with distinctive cell walls impregnated
sible for a fall in the systolic blood pressure and
with silica and containing chlorophyll pigment.
fibrillation.
They are universally distributed in fresh and salt
water and comprise over ten thousand species
(ii) Sea water
ranging in size from a few to several hundred
Unlike fresh water drowning where there is micra in diameter.
haemodilution, in salt water drowning there is The "diatom test" is based on the principle
haemoconcentration. There is withdrawal of that diatom species measuring up to 30 micra in
water from the circulating blood into the lungs diameter are said to be able to enter the pul-
whereby massive pulmonary oedema occurs. monary circulation during immersion and are
Electrolyte exchange from sea water into the disseminated widely to sites like liver, brain,
blood occurs with an increase of plasma sodium kidney and marrow.
levels. Ventricular fibrillation is not a feature
but, with the occurrence of myocardial anoxia Method for demonstrating diatoms
and increased blood viscosity, rapid failure of
Several grams of the organ suspected of
the heart occurs.
harbouring the diatoms (liver, kidney or bone
marrow) are digested with fuming nitric acid
DRo WNING (NEAR until all organic material has been destroyed and
DROWNING)
heating is continued until only a small volume
Persons who survive the initial episode of sub- of fluid remains. The solution is centrifuged
mersion may subsequently suffer complications and the supernatant decanted and discarded.
o r even d i e after a period apparent of The residue is washed several times in dis-
wellbeing.'.'? tilled water. After final centrifugation and dis-
The clinical symptoms and signs include posal of the wash water, the residue is poured on
pyrexia, shallow respiration, blood stained frothy a clean slide, covered with a cover slip and
sputum and pain in the chest with cardiac examined under the microscope with subdued
arrythmias. Electrolyte changes are not signifi- light. The diatom skeletons are readily recog-
cant. Coma precedes death. If active treatment nizable as radially or axially symmetrical
is instituted, the prognosis is good in cases of structures. There are other methods available
fresh water submersion but poor in salt water for the demonstration of diatoms?
immersion.
a. Direct microscopic examinnlion: done only
AQUATIC SPORT AND DROWNING in the examination of the lungs. Water is
squeezed from the lungs and centrifuged
Fatalities associated wtih aquatic sport are usu-
and the sediment is examined.
ally directly attibutable to drowning rather than
hyperbaric effects. In the case of the sub-aqua
diver, the acute problem is one of decompression,
,, Incinerarion the organic matter
(brain, liver etc.) is destroyed with heat in
but the skin diver who deliberately overbreathes
an oven. The principle is that diatoms are
to wash out CO, and thereby increase his breath
heat resistant.
 DROWNING

Interpretation of the diatom test of electrolytes from sea water to the blood also
has its adverse effects. The haematocrit and
The demonstration of diatoms from the sub-
plasma sodium levels rise steeply.
merging fluid in the body of the victim is of
Ventricular fibrillation is not a feature of sea
value as a confirmatory test of death by drowning.
water drowning, and heart failure is slower,
Examinations of lung juices have limited value,
taking from 5-8 minutes. The cause of death is
although a high diatom content is indicative of
myocardial anoxia and the increased viscosity
drowning. T i m p e ~ n a nfound
'~ that after death
of the blood causes heart failure. There is no
diatoms penetrate only as far as the main bronchi.
haemolysis.
Tamaska" regards the identification of diatoms
Modell"' reviewed the mechanism of
in the bone marrow as the best method in all
drowniilg. He suggested that 10% of drowning
circumstances of drowning, including those
victims do not aspirate water but die of asphyxia
where the body is in an advanced state of pu-
due to laryngospasm. He also demonstrated that
trefaction. This view has been confirmed by
when fresh water was aspirated there was a
RushtonIh who, because of the difficulty of total
decrease in serum sodium, chloride and calcium,
exclusion of contamination, considered the
with an increase of potassium in arterial blood.
finding of diatoms as supportive evidence but
Fresh water drowning was considered to be
not as conclusive evidence. W. V. Spitz, at the
twice as lethal as sea water drowning.
Third International Meeting in Forensic Medi-
cine, London in 1963 showed that diatoms were
CHEMICAL TESTS OF DROWNING
plentiful in the air of Berlin.I7
PeabodyL8has therefore suggested the need GettlerZ0and Fisher21 were of the opinion that
for further research to resolve the controversy comparison between the chloride content of
by establishing with certainty whether non- blood samples taken respectively from the right
drowned subjects do have diatoms in their organs and left sides of the heart would furnish a test of
in significant number, and to what extent results death by drowning. This was on the principle
may be vitiated by contamination of reagents that in fresh water drowning, the blood in the
and glassware. left side of the heart, being diluted by water,
should have a lower chloride content than that in
EXPERIMENTAL EVIDENCE the right, and conversely if drowning occurred
in salt water. TimpermanI4 and Model1 and
Experiments of Swan and Spafford19 produced
D a ~ i considered
s ~ ~ these tests untrustworthy.
results on which the modem view of drowning
Rammer and Gerdin2?compared the osmolarity
is based.
and serum sodium and potassium in the left and
right sides of the heart with that in the cerebro-
a. Fresh water
spinal fluid and were of the opinion that a lower
When dogs were submerged in fresh water, osmolarity and a substantially lower concentra-
large amounts of water were rapidly absorbed. tion of sodium and potassium than that found in
Within three minutes the circulating blood could the cerebro-spinal fluid made a diagnosis of
be diluted by as much as 72% with coincident drowning in fresh water highly probable.
haemolysis. The heart muscle thus sustained a For these biochemical tests to be of value as
"biochemical insult". There was an increase in evidence of drowning, samples must be ob-
plasma potassium with a fall in sodium. In tained within a few hours of submersion.
addition, there was myocardial anoxia. Post mortem biochemistry, particularly of
Haemodilution also overloaded the circulation. the vitreous and cerebro-spinal fluid may not
A rapid and considerable fall in systolic blood only be indicative of sea water drowning, but
pressure occurred and within a few minutes, also provide a means of determining how long a
ventricular fibrillation occurred causing cere- dead body may have been in sea water.
bral anoxia and death. Coutseliuis2~emonstrateda progressive in-
crease in magnesium in the vitreous of
b. Sea water exenterated human eyeballs put in sea water and
was of the opinion that this provides a means of
In sea water the mechanism is different. Swan
determining how long a dead body may have
and Spafford showed that in sea water there is been in the sea. The concentration of magnesium
haemoconcentration; up to 42% withdrawal of in the cerebro-spinal fluid helps in establishing
water from the circulating blood into the lungs, the diagnosis of drowning in sea water whilst
causing massive pulmonary oedema. Exchange
Malaysian J Path01 December 1992

magnesium concentration determinations from by other means with subsequent post-mortem

the right and left side of the heart pose problems submersions to stimulate accidental or suicidal
similar to chloride concentration estimations. drowning must always be considered, before
Estimation of serum strontium levels in rabbits being excluded.
before and after death by Abdallah et al.25
concluded that serum strontium levels were a REFERENCES
practical measure for differentiating between I. James WR. A case of drowning in a vat of beer and
death due to drowning, and post mortem immer- two other case reports. Med Sci Law 1966; 6:
sion of a victim. They showed that death by 164-5.
drowning raises the serum strontium levels of 2. Spyker DA. Submersion injury. Epidemiology,
the cadavers, and this was more marked in cases prevention and management. Paed Clin N Am
of sea water drowning than in fresh water 1985; 32(1): 113-25.
3. Pearn JH. Secondary drowning in children. Br
drowning.
Med J 1980; 281: 1103-5.
4. Polson CJ, Gee DJ, Knight B. Essentials of Forensic
ACCIDENT - SUICIDE - HOMICIDE Medicine. 4th ed. Oxford: Pergamon Press, 1985;
421.
Drowning may be accidental, suicidal or homi-
5. Simonin C. Medicine Legalle Judiciare 3rd ed.
cidal. The vast majority are accidentalz6 whilst Paris: Maloine, 1955; 222.
10% to 30%6.26are suicidal. Homicidal drown- 6. Tedeschi CG, Eckert WG, Tedeschi LG. Forensic
ing is only occasionally reported and the victims Medicine. London: WB Saunders. 1977; vol 3:
are usually children. Circumstantial evidence is 1317
of value in assessing whether death was acci- 7. Fisher RS. Harrison's Principles of Internal Medi-
dental - boating accidents, accidents whilst pot cine. loth ed. Tokyo: McGraw Hill, 1985; 754.
8. Gardner E. Med Leg Rev 1942; 10: 120.
holing, or skiing on a lake covered with a layer
9. Spilsbury B. Med Leg Rev 1934; 2: 340.
of thin ice, can result in drowning. Accidental 10. Modell JH. The pathophysiology and treatment of
drownings have occurred in swimming pools, drowning. Acta Anaesth Scand 1968; Suppl. 29:
both in those in use and in disused ones. 263.
Suicidal drowning can occur in similar loca- 11. Donald KW. Drowning. Br Med J 1955; 2: 155-
tions, bath tubs, swimming pools, canals and the 60.
12. Fuller RH. Clinical pathology of human near
sea. When drowning is in shallow water and
drowning. Proc Roy Soc Med 1962; 56: 33-8.
when there are no anatomical or toxicological 13. Mason JK. The Pathology of volent ijury. Nor-
reasons why the victims could not have saved wich, G.B: Fletcher & Son, 1978; 268.
himself or herself (viz. head injury, disabling Timperman J. Medico-legal problems in death by
natural disease or alcoholic intoxication) this drowning. Its diagnosis by the diatom method. J
cause needs serious consideration. Forensic Med 1969; 16: 45-75.
It is not unusual for a would-be suicide by Tamaska L. On diatom demonstration in the bone
marrow of drowned dead bodies. Deutsch Ges
drowning to weight himself or herself with rocks
Gerichtl Med 1961; 51: 398-403.
or heavy weight before entering the water. Rushton DG. Drowning - a review. Med Leg J
Similarly they may tie their hands and/or feet 1961; 29: 90-7.
together before the fatal entry into the water. Spitz W. Diagnosis of death by drowning by means
The nature and type of material and knot used of demonstration of diatoms In organs. Deutsch Z
will be of value in assessing the case. Ges Gerichtl Med 1963; 54: 42-5.
Homicidal drowning may be committed by Peabody AT. Diatoms and drowning - a review.
Med Sci Law 1980; 20(4): 254-61.
an act of omission or commission. The former
Swan HG, Spafford NR. Body salt and water
is occasionally seen in cases of infanticide. The changes during fresh and sea water drowning. Tex
diagnosis of homicidal drowning of children Rep Biol Med 195 1; 9: 356-382.
may be made more difficult by the absence of Gettler AO. A method for the determination of
any other injury. Adult victims of homicidal death by drowning. JAMA 1921; 77: 1650.
drowning on the other hand have injuries indica- Fisher IL. Proceedings of Fourth International
tive of a fight - craniofacial trauma or even Meeting in Forensic Medicine. 1966; 186.
Model JH, Davis JA. Electrolyte changes in hu-
attempted strangulation or evidence of being man drowning victims. Anaesthesiology 1969; 30:
made defenceless by alcohol and drugs. 4 14-20.
Finally in the examination of any body re- Ralilmer L, Gerdin B. Dilution of blood in fresh
covered from water, particularly one where the water drowning. Post-mortem determination of
entry has not been reliably witnessed, homicide osmolarity and electrolytes in blood, cerebro-spinal
fluid and vitreous hunior. Forensic Sci 1976; 8(3):
229-34.
 DROWNING

24. Coutseliuis A, Boukis D. The estimation of mag-

nesium concentration in cerebro-spinal fluid as a
method of drowning diagnosis in sea water. Forenisc
Sci 1976; 7(2): 109- 1 1 .
25. Abdallah AM, Hassan SA, Kabil MA, Ghanim AE.
Serum strontium estimation as a diagnostic criteria
of the type of drowning water. Forensic Sci Int
1985; 28(1): 47-52.
26. Cairns FJ. Deaths from drowning. NZ Med J
1984; 97(4):65-7.