Nasogastric adminsitration of glucose



From An Introduction to Physical Methods of Treatment in Psychiatry (First Edition) by William Sargant and Eliot Slater (1944, Edinburgh, E & S Livingstone).


Feedback about this article?

Not long after insulin came to be generally used in
medicine it was found that in small doses it could be
used to improve appetite in cases of anorexia, and that
there followed a general improvement in the physical and
mental state. The clue was not followed up until Sakel came
to use insulin hypoglycaemia to counter the symptoms of
withdrawal in the treatment of morphine addiction. He
found that he got the best results when a sufficient dosage of
insulin was given to produce clinical hypoglycemia, and that
this phenomenon, hitherto considered dangerous, was readily
controlled by the administration of sugar, when the patients
were watched throughout. He was led to try its use in the
I treatment of schizophrenia, and the Vienna Clinic, which had
seen the first introduction of the malarial treatment of general
paralysis, provided him with the patients to experiment on.
The idea seemed bizarre, but it was found to work. Once
tried its results were, in individual patients, so surprisingly
favourable that its use spread widely and quickly. In the
international congress of 1938 reports were presented of its
trial in most of the civilised countries of the world; nearly all
of them were favourable. The disorganisation of the war years
has to a great extent interrupted its progress in England but
it has gone ahead much faster in America.

Manfred Sakel

Manfred Sakel


Reliable statistics are mostly in favour of the value of the
treatment. In a series of over 400 patients treated in Swiss
hospitals before 1937, when the technique was still new,
59 per cent. of persons treated within the first six months
after onset reached either a complete or social remission. In
the New York State hospitals of over a thousand schizo-
phrenics treated with insulin 11.1 per cent. recovered, 26.5
per cent. made great improvement, and 26 per cent. some
improvement; the corresponding figures in over a thousand
control patients were 3.5 per cent., 11.2 per cent. and 7.4 per
cent. A comparable number of patients treated with cardiazol
convulsion therapy did not do even as well as the control
patients. Furthermore, the treatment greatly shortens the
duration of the illness. Earl Bond reports that in the Penn-
sylvania Hospital 95 per cent, of the patients who were
treated and recovered left the hospital within a month of
termination of treatment; with few exceptions the un-
treated patients who made spontaneous remissions required
one to three years for a similar improvement. Thirty-eight
per cent. of his treated patients remitted, where there had
been previously only 10 per cent, remission and 10 per cent
improvements occurring spontaneously.

The results are much more favourable when treatment is
given early. The New York State hospital figures show that
only 27 per cent. of the patients treated in the first six months
of the illness failed to improve, whereas this figure rose to
66 per cent. for those whose illness had lasted five years
Taking a dividing line of 18 months, Earl Bond finds 67 per
cent, recovery and great improvement with treatment given
before that time after onset, 30 per cent. after. The prog
nostically favourable patients selected for admission to the
Maudsley Hospital showed 34.5 per cent. of social recoveries
when followed up for three years; 70 per cent, of the patients
selected for insulin treatment recovered in the same hospital
at a later time. In both these groups the illness had not
lasted longer than a year.

It is not yet known whether these recoveries will be maintained;
but the three to one ratio in frequency of improvement in treated
as compared with untreated schizophrenics
quoted from the New York State hospitals statistics was still
maintained after two years, and 60 per cent. of the insulin
treated patients were still living in the community mostly
in a recovered or improved condition. Relapses occur both in
patients who have recovered spontaneously and with insulin
It may be that as the years progress treated and control
groups will tend to approximate; but even if this proves
to be true one would wish to keep the potential lunatic sane
as long as possible, and to give him as long a period as
possible of health and happiness. There is also evidence
that the relapsed schizophrenic is still susceptible to
treatment especially if each relapse is treated immediately it

Provided early cases are treated, it is then probably true
to say that insulin treatment brings about a remission quicker
and in a higher proportion of patients than occurs spon-
taneously or with convulsive therapy. There is also a less
unanimous impression that the quality of the remission
obtained with insulin is better than that of the spontaneous
remission; this is likely to be true, if for no other reason than
that recovery takes place earlier under treatment and there is
less time for the psychological scarring that is the most
terrible effect of the disease. There is unanimity that in the
earliest months of the illness the results are out of all pro-
portion better than later on. In some patients the illness
comes on so insidiously that even retrospectively none but
the widest limits can be assigned to the time when it started,
and these patients it is generally agreed are the most un-
favourable therapeutically. This may quite possibly be not
because the form of the illness is more malignant—ordinary
clinical judgment suggests the opposite conclusion—but
because in all these patients the illness is likely~ to have lasted
for months or even years before it is even suspected.

Finally, it is clear that there is an art of treatment, in which
some will be more adept than others, and that operations
cannot be conducted by rule of thumb. Different workers
have obtained very different results with the same type of
material, both in the frequency of remission and of unpleasant
complications. Earl Bond notes that in a group treated in
the Pennsylvania Hospital between 1936 and 1938 treatment
was given tentatively and was in the hands of several different
physicians and all possible risks were avoided; only 46 per
cent. of remissions was obtained in cases of less than a year’s
duration. After 1938 a single skilled therapist was employed,
and the comparable remission rate rose to 79 per cent. This
may be partly because where there is a specific treatment
available, patients still in the early stages of the illness are
encouraged to seek treatment; but at least part of the im-
provement in the recovery rate must be attributed to the
greater expertness of the treatment and the giving of really
deep comas when necessary. It is only too easy to carry out
the treatment in a slipshod or incautious way, or to err on the
side of over-easy discouragement. The successful therapist
will be gifted with enthusiasm and caution, he will have a
sympathetic interest in and a detached appreciation of the
personalities of his patients, he will have the general medical
training that has accustomed him to the handling of medical
emergencies, and the refined clinical judgment of the èx-
perienced psychiatrist, and he will have the facilities to give
the whole of his energy to the treatment of his patients
without administrative after-thought. The abilities of the
therapist are not less important than the method of treat-
ment adopted. On his skill alone the recovery of any particular
patient may ultimately depend.


It is rarely indeed that facilities will exist for the treatment
by a full course of insulin of all the schizophrenics coming
under observation, and it is therefore important not to waste
the treatment on patients not very likely to respond while
denying it to the favourable case. The first point for considera-
tion is the length of duration of the illness, and this does not
mean of overt symptoms. The patient who has only recently
come to show definite and unmistakable symptoms but has
been known to have been gradually becoming queerer for
several years is not a favourable case for treatment. The most
favourable case would be the patient who had been well up
to a few days or weeks of being seen, and he should be selected
even if there is some lingering doubt of the true nature of the
illness and probabilities only speak in favour of schizophrenia.
An atypical onset should not be allowed to develop gradually
over the course of months into an unmistakable clinical
picture before treatment is begun. This position is likely to
arise in atypical manic excitements, in obscure confusional
states, and even in depressive states with a suggestion of
catatonia, of hallucinations or other suggestive symptoms. In
any case it must be remembered that schizophrenic psychoses
greatly outnumber all others in persons under the age of
thirty. From the much greater success in the early case it
follows that once treatment is decided on it should be begun
at the earliest possible moment. If there is a waiting list for
admission, the early schizophrenic should take precedence of
most other patients, firstly for the reason already given, and
secondly because in the early stages the clinical picture is
very fluid and an attempted suicide, an attack of excitement
or violence may remove the patient from the possibility of
treatment where it has been arranged to another place where
there may be further delays. As even a few weeks makes a
difference, this would be a serious matter.
The rapidity with which treatment can be inaugurated
after the onset of the illness is by far the most important
factor therapeutically. Next in importance ranks the quality
of the personality before the illness began. A frank, open and
socially well-adjusted personality reacts better than one
which has always been shy, shut off, awkward and autistic.
There are probably a number of reasons for this. The autistic
personality may not have always been so, but have become
so as the result of an early and unrecognised schizophrenic
process, which -has perhaps become chronic, only to show a
recent florid exacerbation. Further, restoration cannot at
its optimum be to the same level in the defective as in the
thoroughly normal personality. This is perhaps the principal
factor that militates against the successful treatment of
schizophrenia in the intellectually retarded.
Rapidity of onset is generally held to be a favourable factor,
which would imply that in two schizophrenics in both of
whom the illness was of three months' duration, the one in
whom it began in florid form would have the advantage over
the other in whom it began gradually. This may be because
in the latter the true point of onset would be set further back ;
but it may also be because the more acute illness is more easily
The bodily physique has been found to be of importance,
and the pyknic or athletic habitus is more favourab!e than the
asthenic and the dysplastic. This is a very general clinical
impression, and has recently been confirmed by Freudenberg.
Freudenberg has also found that an abundance of " process
symptoms," such as hallucinations, thought disorder, primary
delusions, passivity feelings, etc., are unfavourable. A history
of a previous attack with a full remission, is on the other hand,
a favourable sign. It is also our impression that an atypical
quality in the symptomatology is a sign of good omen.
These favourable and unfavourable factors seem to act
cumulatively, and very satisfactory results can be expected
from the patient who shows all of the first and none of the
second, whereas if the opposite is true treatment is hardly
worth while. In patients who fall between these two groups
the results obtained will also be intermediate; there will not
be the same frequency of complete success as in the most
favourable group, but considerable improvement, sufficient
for social rehabilitation, will often be obtained even where a
complete remission appears unattainable. It will be seen that
the factors that favour a satisfactory response to insulin are
also those clinically associated with a higher expectation of
spontaneous remission; but this would be an inadequate
argument in favour of a laissez-faire attitude. The tragedies of
neglected insulin treatment in England are to-day a commonplace
to the psychiatrist of experience; we have as yet
seen no tragedies from premature treatment skilfully applied.


The risks of treatment are in general less than the risks of
waiting for spontaneous remission to occur. Massed figures
gathered in the United States give a mortality of 90 deaths in
12,000 patients treated, of which about half were due to
hypoglycaemic encephalopathy, which is thus seen to be the
most serious risk of treatment; it is, on the other hand, a:n
avoidable risk and is the rarer the more skilled the operator.
In this series there were also twelve deaths from heart failure,
nine from aspiration pneumonia, seven from pneumonia
occurring otherwise, some of which were probably also avoid-
able. The New York State hospital figures show that there was
a higher mortality in the untreated than in the treated
patients. It may be that patients with better physique were
selected for treatment, and if so that could partly explain
the result. It is also true that the patient's bodily health
rapidly deteriorates. under an acute schizophrenic psychosis,
thereby laying him open to a greater risk of tuberculosis and
intercurrent disease .han where the process is cut short by
treatment. Twelve patients died of pulmonary tuberculosis
in the control group, only one in the insulin treated group. A
mortality of less than one per cent. cannot be considered a
reasonable bar to treatment in the average patient; and the
risk of death or serious damage to bodily health can be
neglected in all but the physically ill. Any advanced degree
of cardiac disease is usually considered a contraindication to
treatment, as are also Graves’ disease, diabetes, liver and
kidney diseases causing marked impairment of function.
Treatment can be dangerous, and is rarely successful, above the
age of forty-five.
Careful investigations have shown that in patients with a
prolonged series of deep comas there is sometimes a mild
degree of intellectual impairment the effect is much smaller
than that also seen in patients of all kinds treated by con-
vulsion therapy, which is far from great or constant. The
degree of impairment has been of practical importance in
only a handful of patients reported in the literature, and is in
any case not comparable with the disability caused by the
disease itself. Mental impairment is of much greater im-
portance after long irreversible coma, and a severe Korsakow
picture can result from this. This improves somewhat as a
rule, but usually leaves a greater or lesser degree of permanent
impairment. Its occurrence is the principal reason why the
deliberate production of irreversible comas, which often have
an immediate curative effect on the schizophrenic process, has
not been extensively tried as a method of treatment, and
should remain exclusively in the hands of treatment experts
of long experience. A minor but still important reason why
treatment as early as possible is so desirable, is that there is a
lesser necessity for repeated deep comas, with the attendant
risk of irreversible coma.


We are indebted to the Journai of Mental Science for permission to
include in this chapter the illustration and some of the details of technique
from an article by Fraser and Sargent 1940

The beginner would be well advised, before starting this
treatment, to go for a week or more if possible to a hospital
where it is done on a fair scale. The technique cannot .be
learned from books alone, and some practical experience is
necessary of the dangers that may be met with.
The following technique here described mostly arises from
experience gained from 1937 onwards at the Maudsley Hospital
in conjunction with Dr. Russell Fraser, and subsequently in
an E.M.S. Neurological Unit where early cases could be
admitted and treated under the conditions obtaining in a
general hospital.
Lack of the best facilities should not prevent treatment
altogether; and the dangers of treatment under adverse cir-
cumstances may have to be balanced against the dangers of
delay. It is quite possible to carry out the insulin treatment
of schizophrenia in a general ward, with the patient screened
off during the coma period. Nevertheless where schizophrenics
are going to be treated in any number a special insulin
treatment unit is very desirable. Excited patients cannot be
easily handled in general wards, and even quietly behaved
schizophrenics are likely to become noisy and difficult during
hypoglycaemia. A fairly large room should be taken, in which
all the patients it is proposed to treat at one time can be kept
economically under observation: off this there should be
one or two side-rooms available for the more restless and
noisy. These rooms will be kept at a warm temperature, as
patients perspire profusely and often throw off their bedclothes.
To staff the unit a well-trained team of nurses is required,
of whom as many as possible should have both a ‘general
and a psychiatric training. Experience of medical and
surgical emergencies is very valuable in the management of
this complicated method of treatment. A sister should be
in charge, and on duty during the treatment period every
day. Changes in staff from day to day mean that minor
alterations in the behaviour of individual patients during
sopor and coma may be missed, to the danger of the
patients: for behaviour varies widely from patient to patient
while remaining very constant to the individual, and a nurse
who knows the patient well can detect signs of danger Jhat
would not be at all noticeable to anyone else. Every effort
should therefore be made to keep the team together, and avoid
changes of more than one of the personnel at a time or at
less than well-spaced intervals. The sister is of coutse the
most important of all; and it will be her responsibility to
assemble and maintain the equipment so that none is astray
at an emergency.
The patients also should be kept together through the
greater part of the day. Their food intake at every meal has
to be carefully supervised, and they themselves must be kept
under supervision to prevent the occurrence of after-shocks
in the afternoon or evening. If they go out in the afternoon,
it should be in the company of a nurse supplied with glucose
to give to any of them who may suddenly develop a recurrence
of hypoglycemic symptoms.
Apart from the usual ward apparatus such as intravenous
syringes (1 to 20 c.c.), intravenous needles kept freshly
sharpened and in perfect state, a variety of small basins,
swabs, surgical spirit, etc., the following special equipment
is required: blood-pressure apparatus, air-way, tongue-clips,
mouth gag for use in the event of a fit (a rubber heel covered
in mackintosh or a dog’s small rubber bone are serviceable),
an emergency apparatus for giving 5 per cent. carbon dioxide
in oxygen, ampoules of adrenalin, morphine, hyoscine, 10 per
cent. calcium chloride, sodium amytal, atropine, coramine,
eucortone, nicotinic acid, vit. B.1, and luminal (all suitable
for intravenous and intramuscular use). The means of cutting
down on a vein should also be at hand.
On one or two small portable trays are placed the means of
nasal interruption. The nasal tubes selected should be stiff and
of fine bore, and should be discarded when they get soft from
repeated boiling. On the same trays there will be lubricating
oil, litmus paper and an aspiration syringe for sucking out a
test sample of the stomach contents through the nasal tube.
The container of 600 c.c. of 38 per cent. sugared tea (6.5 oz.
sugar to the pint) is also on this tray. The tea should be
prepared before treatment starts in the morning and kept till
needed in bulk in a large jug containing sufficient for the
needs of all patients. This reservoir of sugared tea must be
kept warm, for instance in a large thermos jug, and from it
the smaller receptacles for individual trays can be filled just
before interruption.
Another special tray or trolley holds the equipment for an
emergency intravenous interruption—one or two syringes each
containing 20 c.c. of fresh sterile 83 per cent. glucose solution
and a sterilised bowl from which they may be rapidly refilled.
Sealed bottles of 38 per cent. glucose should also be at hand,
and only opened before the injection starts. There should be
ample amounts of 5 per cent. glucose saline ready for emer-
gency intravenous use, but it need not be specially heated.

Sarganta and Fraser have adapted for use in insulin therapy a
composite pressure apparatus which avoids the necssity of having
to change syringes in giving large quantities of 33 per cent glucose
or glucose salines intravenously, with the risk of losing one's place in
the vein every time the change is made. From the illustration it will
be seen that it is a simple pressure bottle attached to a syringe with a
side-valve. The needle is inserted into the vein, and with the first part of the

Glucose solution (33%) for reversing coma

plunger, blood is sucked into the syringe, thereby showing that the needle is
in position. Further withdrawal of the plunger beyond the side-valve allows
the 33 per cent glucose solution in the pressure bottle to flow freely through
the syringe into the vein. Bottles of 5 per cent saline are kept in stock which
also fit the same screwcap of the apparatus. If after the 33 per cent glucose
has been given, a change to 5 per cent glucose or saline is required, this is
effected simply by unscrewing one bottle and screwing on the other, pressure
being re-established by more hand pumping. Through all this, the doctor's
attention can be concentrated on the more important job of keeping the needle
in place in the vein, a ticklish matter if the patient is excited and restless, while
the nurse does the pumping and
changes the bottles if .necessary. Emergency drugs such as
coramine can be injected into the rubber connection between
the bottle and the side-valve syringe without further interference with the veins.
For the avoidance of unnecessary risks, the closest atten-
tion to record keeping is required. The quality of the records
will often distinguish a good clinic from a careless one. Four
separate records are recommended. There should be a tem-
perature chart,
on which are entered the temperature, pulse
and respiration rate daily at 6 a.m. and 9 p.m., the diet taken
(as “full,” “half,” or “excess”), any tube feed that may have
had to be given, the daily insulin dose, duration of coma and
time taken to awake after interruption. Any special com-
plication is marked by an asterisk on the chart. The weight
recorded weekly or hi-weekly is also noted. A space is left at
the side of the chart for entering the average coma dose of
insulin and the maximum safe period of coma for that par-
ticular patient, after these have been ascertained during the
A daily treatment chart is also kept. On it are shown half-
hourly records of pulse and respiration rate during treatment,
the amount of sweating, the amounts of insulin and glucose
given each day, the time of onset of sopor and coma and the
time taken to come round after interruption. Details of the
patient’s behaviour and neurological abnormalities in sopor
and coma are also noted here.
In a conspicuous and convenient position in the ward there
is a treatment board. On it are marked the times of onset of
sopor and coma, the time when each patient is due to be in-
terrupted, the time he is actually interrupted and the time
of his awakening. It is filled in as these events occur by the
doctor and nurses. The important data it provides can be read
at a glance from any part of the ward, and help to prevent
delays and omissions when emergencies engage the attention
of the staff too exclusively to one patient. It is also useful for
providing the data of the more detailed and permanent records.
Lastly, there should be an insulin dosage book, in which the
next day’s doses are entered at the end of each morning’s
work. It is for the information of the sister or nurse in charge
when giving the insulin the next morning.

Before treatment is begun, it is desirable to deal with any
septic foci that may be present, at least if of gross degree and
fairly easily treated. There is evidence that insulin resistance
is increased by even a mild infective process.
Great importance should be placed on giving a correctly
balanced diet during treatment. If possible, measured normal
meals should be prescribed and kept approximately of the
same content each day. Excess of carbohydrates in the diet
enhances the probability of after-shocks by increasing insulin
sensitivity. The aim should be to have every patient on a
constant diet, a regular daily routine and the same length of
coma each morning. When the metabolism is kept steady in
this way, it is rare to get sudden changes in the response
to the morning insulin or the occurrence of after-shock,
Furthermore, any ominous irregularity in behaviour is
more likely to be noticed and reported. If the diet taken is
inadequate tube-feeding must be resorted to. It must be re-
membered that a morning dose of insulin may still be in process
of absorption for eighteen hours after it has been injected.
Much of the treatment devolves on the nurse, and on her
skill its success depends. Provided she is well chosen, the
more responsibility she is given the better the standard of
work she will attain. The sister and the senior nurses should
be trained in all details of the technique, from the passing of
the stomach-tube to the intravenous interruption, and they
should be able to carry out these procedures with confidence.
They learn to judge for themselves when the patient is going
too deep, or showing abnormal symptoms, and to report to the
doctor in time. In an emergency they are taught to act for
themselves, and may have to do so if complications arise in
several patients at a time and the doctor’s whole attention
has to be concentrated on only one. If the selected male and
female nursing staff are trained to a high pitch of efficiency,
a greater number of patients can be treated than if the doctor
has to undertake every routine procedure himself. A doctor
must always be on call, but it is not necessary with a trained
staff for him to be in the insulin room except during the
times of sopor and coma. If the nurses have been especially
well selected and trained they can even be left to handle the
early stages of sopor and light coma provided the doctor can
get to the treatment ward within a minute of being summoned.

We give full treatment on five days a week. On the sixth
half-doses of insulin are given, and the seventh is a rest day.
If time is short, however, full treatment can be advantage-
ously given on six days a week. The commencing dose of
insulin is usually 20 units at 7 a.m. given intramuscularly by
the nurse in charge, with the patient fasting from 8 p.m. the
night before, and the dose is increased by this amount each
day of full treatment until sopor occurs, when progression of
dosage may be slower. When high doses of insulin have been
reached without sopor, the jump may be 80 to 40 units each
morning instead of 20 units. Once coma has been induced,
the insulin dose should be adjusted until the minimum
satisfactory coma dose has been found. Neglect of this
principle will lead to the occurrence of irreversible comas.
Insulin sensitivity frequently increases during treatment,
and the regular daily dose may eventually be stabilised at
half the amount that was necessary to produce the first coma.
Furthermore, often enough the patient seems consciously to
resist going into sopor or coma, and when he finally does
so a much lower dose of insulin is needed to induce the same
state on subsequent occasions. Sometimes sodium amytal
gr. 3 to 6 has been given by mouth at 7 a.m. when the insulin
is given. This may make the patient less restless and enable
him to pass more smoothly into sopor. This dose of amytal
is not enough to make the sleep induced by the drug indis-
tinguishable from the hypoglycemic phenomena. Sometimes
very high doses of insulin fail to produce conih at all. Then
the doses must be “swung.” On one day 240 units should
be given, and then in rotation 40, 120 and 240 units. In this
way insulin sensitivity may often be increased to the point
where a satisfactory coma is induced. Insulin resistance up
to 600 units has, however, been reported, and the phenomena
of insulin resistance are little understood. Some patients, for
instance, may go into coma with as little as 40 units or even
There is much confusion about the correct use of the word
“coma". It is generally agreed that the length of time the
patient is actually unconscious is the best index of the severity
of a period of hypoglycemia. Yet in papers on technique

daily “comas” of two and a half hours are recommended by
some, while others advocate only three-quarters of an hour.
These discrepancies are due to differences in the meaning given
to the word coma. The criteria of coma given by Kuppers
have been adopted in our work. He differentiates the stages
of loss of consciousness into sopor or pre-coma and true coma.
Because of individual variations it is generally found that
reflexes, motor phenomena, and most other physical signs
cannot be used as criteria of onset of either of these stages,
and reliance has to be placed on tests for the presence or
absence of “ conscious “ or purposive reactions.
The onset of sopor is usually indicated by the loss of a
normal response to speech and impairment of orientation.
But testing will still elicit Ssome confused but purposive re-
sponses. There may be some difficulty in deciding the exact
time of onset of sopor from earlier degrees of hypoglyc~mia.
This is not so important as to recognise when sopor deepens
into coma. The onset of coma is distinguished by the loss of
all purposive responses, simulating those of the conscious
patient, even on careful testing. There should be no responses
from visual, auditory or tactile stimuli. Painful stimuli may
still produce some movements, but these are not directed
towards the stimulus., Tests should include raising and
dropping the patient’s arm, trying to make his eyes follow
a moving object, and giving a painful stimulus such as pressure
on the supraorbital nerve. As coma supervenes the eyes may
still remain open and some non-purposive movements persist,
but the absence of purposive response can always be demon-
strated by testing. Some patients, while still in sopor and
aware of the test stimuli, lose all initiative to respond. They
must be distinguished from those in coma, and generally
painful stimuli will reveal the difference.


The second phase of treatment starts when the patient
begins to go into sopor. The regulation of hypoglycaemia to
achieve the maximum degree of safety is best done by con-
centrating on the duration of actual coma rather than of
sopor. But very occasionally hypoglycaemia has become
"irreversible” when the patient is allowed to remain at the
stage of apparent sopor for a very long tIme. As a precaution,
therefore, interruption is carried out after an hour and a half
of sopor, if coma has not supervened.

When the patient has begun to go into coma the length of
the coma period allowed is increased gradually from five
minutes on the first day to what proves to be the maximum
safe duration for the individual. Using the criteria given
above, the average daily period is half an hour, with con-
siderable individual variations above and below this figure.
The physician should not rely on rule of thumb, but should
try to discover the safety limit for each patient. He may be
warned that coma is getting too deep by the patient taking
over twenty minutes to awake after the nasal feed. He should
be guided by this, or interrupt when other signs of excessive
depth occur. As the patient’s physique improves with treat-
ment, longer comas can be tolerated than initially; this will
be indicated by changes in the depth of coma or alteration in
waking time. Sometimes it may be desired to take the patient
particularly deep. If this is done it is wise to interrupt
immediately afterwards by the intravenous route, and not to
wait a further twenty minutes or less for the patient to come
round after a nasal interruption—unless after thorough
testing this has been found to be safe. A particularly severe
coma is also apt to cause an increased susceptibility to the
next day’s insulin, and this must be remembered if deep
comas are given. To prevent irreversible coma, a shorter
coma is advised on the day following a deep coma, particularly
if there have been signs of shock or delayed awakening. For
the general run of patients, it is best to give a maximum safe
coma treatment, and to stick to it each day until recovery is
manifest. Some patients, however, do well with submaximal
comas, and others need dangerously deep comas to achieve
results. This is why results vary from therapist to therapist,
and success depends so much on his skill. With increas-
ing depth of coma comes an added risk of irreversible
coma, and great skill is needed to handle this emergency,
but the risk must be faced if it is necessary to the patient’s

Our remarks apply to comas occurring at the end of the
third and the beginning of the fourth hours, and are not
applicable to those beginning in the fifth hour. Margins of

safety must necessarily be loWer in the latter. The insulin
dosage should be so adjusted as to induce coma about three
hours after injection.


Signs of excessive depth call for interruption at any stage
of coma. The peripheral circulation and blood pressure are
valuable indicators of circulatory embarrassment, and re-
peated examination of the finger tips is advisable. When the
blood pressure falls below 100mm. or the peripheral circulation
becomes poor the coma should usually be interrupted. A
falling blood pressure and a rising pulse or respiration rate
should always be regarded as a danger sign, especially if
combined with signs of failure of peripheral circulation.
Interruption must be done in good time; the tube feed may
be vomited or poorly absorbed, and when intravenous inter-
ruption is attempted the veins may be found to be collapsed
and may then have to be cut down on. If pulse irregularities
first appear during the later stages of coma and the pulse drops
below 55, interruption is advisable. Earlier pulse irregularities
before the onset of coma often subside with its onset, and
should only require caution when they are frequent or persist
for over half an hour. As a general rule, provided the blood
pressure remains above 100, the pulse volume is good and the
pulse between 70 and 100 per minute, these patients may be
left for three-quarters of an hour before interruption. Some
patients who start with extra-systoles in the initial stages of
treatment lose them as treatment progresses.
Motor neurological signs are of little help in determining
the onset of coma, but they are useful indications of its depth.
In the earlier phase various types of movement may appear;
they are clinically important only if they are excessive and
produce exhaustion. Occasionally they may be the pre-
monitory signs of an impending fit. The movements should
not be allowed to continue for over an hour and a half or less
if the patient is becoming exhausted by them. Premedication
with luminal gr. 1 to 2 or sodium amytal gr. 8 to 6 may help to
diminish them or prevent their occurrence. If the jerking and
movements are too great it is sometimes advisable to give
intravenous sodium amytal gr. 2 to 3.5 at the time to reduce them.

The drowsiness induced by such a small dose of intravenous
amytal is easily distinguished from true coma and may
enable the patient to pass through the restless phase into

According to Mayer-Gross severe hyperkinetic conditions,
restlessness and noisy excitement, can also be controlled by
giving a part or the whole of the morning insulin dose intra-
venously. This brings on the deeper and quieter stages of
coma more quickly. One has to learn by trial in each case
how much of the total dose to give intramuscularly early on,
and how much intravenously at a later stage in the day’s

In the deeper stages of coma waves of extensor tonus
occur which are well seen in the arms as combined extension
and pronation. They are really important, as they indicate
that the safe limit of coma is being reached. But these move-
ments are often precipitated or exaggerated by respiratory
embarrassment or circulatory failure. If the air passages are
freed by inserting a Hewlett’s airway and the extensor tonus
then subsides, coma may be allowed to continue. When these
waves of extensor tonus are only spasmodic and the state of
the circulation is satisfactory, coma may be continued for a
further fifteen minutes before interruption. But to be on the
safe side it may be advisable to interrupt intravenously at the
end of this time and not await the slower operation of a nasal
feed. If the waves persist for longer than a minute, or the
circulation is poor, immediate interruption should be done.
Generalised tremor when the patient is not cold is another
important sign of excessive depth and calls for early inter-
ruption. After going into coma sometimes the patient
awakens spontaneously, generally after a period of severe
spasmodic movements. If he is allowed to relapse into coma
agail after this it should oniy be for half the normal length of
coma, or even less, as this phenomenon may be the precursor
of an irreversible coma. The condition of the reflexes is of
little help in giving warning of excessive depth. More im-
portant is any change from the usual ueurological pattern
seen during previous comas. If unconsciousness seems deeper
or the neurological pattern different from the usual for that
particular patient, and the circulation is poor, interruption
should be done as a precautionary measure.


Epileptic fits occur early or late in the hypoglycaemic
period. The early fit occurs 45 to 100 minutes after the start
of treatment and before the onset of coma. It is generally
easy to manage. Often after it is over the patient wakes up
spontaneously and can drink his sugared tea, or glucose can
be given nasally or intravenously if he remains confused.
Fits occurring in late sopor or during coma are more dangerous
and may be followed, especially in the case of those occurring
in the later stages of coma, by delayed recovery or severe
shock. Immediate intravenous interruption is necessary for
these later fits. Sometimes absorption of sugar from the
stomach does not occur for some hours afterwards, and
therefore further intravenous glucose may be necessary in an
hour’s time, and even again later in the day. Fluids are
valuable when signs of shock are present; up to 500 cc. of
5 per cent. glucose saline may be given after an initial 250 cc.
of 33 per cent. glucose intravenously. Late fits in the stage
of. coma probably indicate excessive cerebral glycopenia,
and that is why they should be dealt with efficiently and

Respiratory complications occasionally arise from aspiration
of saliva during coma, because of increase of salivation and
diminution of protective reflexes. In the later stages of sopor
and during coma the patient is best turned on his side to
permit the saliva to drool out of the mouth. Respiratory
stridor occurring in coma and not due to blocking of the air
passages nor relieved by passage of an airway calls for im-
mediate intravenous interruption as it may be a dangerous
Vomiting after nasal interruption can often be dealt with by
giving atropine gr .001 to .02 intramuscularly some minutes
before the nasal feed. It is only a dangerous complication
when it indicates a condition of shock. If it does, the treat-
ment of shock will be required, i.e. intravenous glucose and,
if necessary, saline.


In the lightest stages of hypoglycemia the patient can, as
a rule, drink 600 c.c. of 33 per cent. sugared tea (for preference

flavoured with lemon) at the end of treatment. In sopor and
coma the same amount has to be given by means of a nasal
tube. When the tube has been passed, gastric juice should
be withdrawn and tested for acidity with litmus paper before
the sugar is poured down the tube. When the gastric juice
is only weakly acid a teaspoonful of salt is added in case
chloride deficiency is being caused by excessive sweating.
When coma has been deep it is unwise to allow more than
twenty minutes to elapse in waiting for the patient to come
round before giving glucose intravenously. This time may
have to be shortened if coma remains deep or other signs of
danger are present. When a skilful technique in the pre-
servation of the veins and the use of the pressure bottle has
been acquired, therapy may be conducted on the basis of a
routine intravenous interruption. Longer and deeper comas
can then be given with greater safety. The state of the veins
must be kept under observation and one or two unharmed
veins should be always kept in reserve for use in emergency;
if thrombosis has occurred in most of the conveniently placed
veins a return to nasal interruptions is indicated. In inter-
ruption by the intravenous route 100 c.c. of 83 per cent.
glucose is generally sufficient, but up to 250 c.c. may be given
with safety. The patient receives further sugar to drink
when he awakes.

After the patient has come round he may be given a light
breakfast. Lunch should be given about 1 p.m., tea at 4.30, and
another meal in the evening. The commonest time for after
shocks is about four hours after interruption. If the patient
is not taking proper meals additional glucose may be given
at this time as a prophylactic.. But it has already been
pointed out that too great a carbohydrate preponderance in
the diet is undesirable. Regular balanced supervised meals
at the right times and a correct morning dose of insulin are
the best prevention of after-shocks.


This is a dangerous complication in which unconsciousness
persists despite the giving of adequate intravenous glucose. In
severe cases there are pronounced vascular shock, hypertonus,
writhing movements, and the condition resembles one of

anoxia. The milder cases may merely show delayed local
recovery, such as monoplegia or aphasia or a slight confusional
state of short duration. The severe case may involve days of
unconsciousness, but with skilful handling death should be
avoided. The pathology of the condition is unknown; but
treatment must be directed towards facilitating the entry of
glucose into the brain cells, combating the alkalosis and
anoxia, and restoration of the circulation to its maximal
efficiency. We have seen great benefit from the giving of large
amounts of intravenous salines in these cases. When the
patient does not come round as he should with the intra-
venous administration of 100 c.c. of 38 per cent, glucose, this
condition must be assumed to be present; and before the
needle is withdrawn an additional 150 c.c. of 33 per cent.
glucose should be immediately given. If the patient still has
not come round, 10 to 80 c.c. of 10 per cent. calcium chloride,
2 to 4 c.c. of coramine, 20 mgms. of vitamin B. 1, and 500 to
1000 c.c. of 5 per cent, glucose in saline should be given intra-
venously, starting as soon as possible. The end of the bed
should be raised, 5 per cent. CO2 in oxygen is given to the
patient and warmth applied, including hot-water bottles, as
for shock. After these measures the degree of shock should
be subsiding, and the rectal temperature will generally rise,
but the patient is often still very restless. An injection of
morphia and hyoscine is useful for allaying this.

An hour after the first injection a second injection of
250 c.c. of 33 per cent. glucose, together with a further 500 c.c.
of saline, should be given if necessary. If treatment has been
prompt and efficient, by this time the danger of acute collapse
is usually past; but the patient may not recover consciousness
for many hours or even days. Often absorption from the
stomach starts again very slowly, and further intravenous
glucose has to be given later in the day, as insulin is still being
absorbed from the morning Thjection. A stomach-tube
should be passed when first aid measures have been completed,
and four-hourly nasal feeds of glucose are given. If these are
not being absorbed they are withdrawn every four hours and
fresh glucose substituted until stomach absorption restarts.
Each time the previous feed is found unabsorbed, or has been
vomited, 500 c.c. of 5 per cent, glucose in saline should be
given intravenously. In very severe cases in which it does not
appear probable that the patient will come round tor a long
time a vein should be cut down on and a continuous glucose
saline drip started. Blood transfusions may be given alternat-
ing with the saline drip. Eucortone has been recommended.
Nicotinic acid mgms. 100 four-hourly may also be given
during the period of unconsciousness by the intravenous
route via the continuous drip. The patient’s strength must
be maintained by intravenous feeding till stomach absorption

With prompt treatment recovery from this condition gener-
ally takes place in the first twenty-four hours; but a period of
unconsciousness lasting six days has been seen, when the
doctor had failed to institute emergency measures immediately.
After long periods of unconsciousness very marked impair-
ment of intellectual functions will be seen, but a surprising
degree of recovery from this will occur in succeeding weeks
and months. Following short-lived irreversible comas some
dramatic improvements in schizophrenic symptomatology
have been recorded. If, with effective measures, the patient
has emerged rapidly from irreversible coma, the fact that the
complication has occurred does not imply that treatment
should be abandoned. It may be started again in three or
four days’ time, but thereafter the comas must be kept much
shorter. On several occasions we have known treatment,
started again under these circumstances, to proceed without
further mischance to a successful conclusion.


In patients who are going to react well to the treatment it
will generally be found that after the first few comas, wakening
from coma leads to an hour or two of a considerably improved
mental state. Sometimes in the early case improvement
occurs without the coma stage having been reached. The
most prominent change is a great improvement in rapport and
the affective attitude. For a time the patient regards the
doctor and the nurses in a much warmer and more friendly
manner, suspicion is for the moment in abeyance, and there is
often a considerable degree of insight into the delusions, feel-
ings of unreality, etc. After this short interval, however, the
patient sinks back into his old state, and remains in it for the
rest of the day. Patients who are excited, agitated or panicky
are often much improved in this respect for a longer period of
time. As the days of treatment succeed one another the
beneficial change lasts longer, and the state into which the
patient relapses becomes less hind less severe, until gradually
the improvement is maximal. In the most fortunate patients
this may occur with astonishing rapidity, less than a fortnight
of intensive therapy sufficing to bring about what appears to
be a complete cure.

In order to judge whether the maximum amount of benefit
has been gained it is necessary to have a very thorough
knowledge of the patient’s symptoms and clinical state before
beginning treatment. As improvement takes place delusions
and other morbid ideas which have hitherto been concealed
may be brought to the light of day, and give a false im-
pression to the naive that the state is becoming more acute.
A very careful clinical examination should be repeated when
the termination of treatment is being considered; and treat-
ment should not be stopped if there are any signs of activity
of the disease, unless it is thought hopeless to proceed any’
further. By this it is not meant that the patient must have
full insight into the symptoms of the past, but that there
should not still be hallucinations occurring, or feelings of
influence or passivity, etc.

Furthermore, it is probably desirable to continue a more
modified form of insulin treatment until the patient has
regained the physique that is normal and healthy for him.
It is quite possible that relapse is more likely, even when there
is complete restoration of mental normality, if a really good
state of bodily health has not been regained. Mental im-
provement and physical improvement usually go side by side,
but one may lag a bit behind the other. If the patient regains
his normal weight and physical state, and retains his mental
symptoms unaltered, it is a bad sign, and it is usually not worth
proceeding much further. Most favourable cases respond to
treatment within two months. Sometimes recovery seems to’
occur two to four weeks after treatment has been completed.
If no change has been brought about in three months, and
every therapeutic trick has been tried, it is unlikely that there
will be a favourable outcome,




Although convulsive therapy is no method of treatment of
the schizophrenic psychosis itself, it can often play an adjuvant
part, and be particularly useful for dealing with individual
symptoms. Depression in schizophrenia, as in basically
different disorders, is often much improved by a few electrical
fits; and their use before the treatment proper is begun may
allow the patient to be more easily treated, and may, if neces-
sary, permit the physician to get a clearer perception of how
much schizophrenic disturbance there is actually present.

Depression in schizophrenia is a secondary symptom,
possibly partly psychogenic in origin, possibly partly a result
of the metabolic changes that are having a total effect on the
patient’s health; but the intensity of the affective change
may obscure the realisation of the more fundamental and
ominous symptoms. The hypothymia, lassitude and anergia,
which are more directly related to the schizophrenic process
may also clear up under the influence of a few convulsions,
which are more likely to be required for that purpose after the
termination of insulin treatment than before it has begun.

Catatonic stupor usually yields to convulsive therapy, but
sometimes passes into a catatonic excitement.
Convulsions should then be used if there is doubt of the
diagnosis, or if the catatonic state interferes with the practical
details of insulin therapy, e.g. in the intake of an adequate
diet, Qr if the condition fails to respond fairly rapidly to
insulin and a depressive component is suspected. The insulin
treatment will itself be nearly always required to establish
ground gained by convulsive therapy, and to prevent the
possibility of rapid relapse. Insulin will also remedy sympl~ms
of thought disorder untouched by convulsions. There is
some evidence that a few convulsions may be of benefit
during the course of insulin treatment, for instance when the
patient has started to improve, but has failed to maintain
improvement. Where the two treatments are combined they
should be given separately and not on the same day. The
role of convulsive therapy in schizophrenia is therefore a
supplementary but important one; the improvements claimed
in the past from convulsions alone have mostly been symp-
tomatic ones without significance for the course of the
disease, or have proved temporary. Some of the few apparent
cures may well have occurred in what were fundamentally
depressive illnesses, with a schizoid colouring derived from the
structure of the personality. The highest recovery rate in
any large and varied group of cases .initiaiiy diagnosed as
schizophrenia will be achieved when insulin and convulsion
therapy are skilfully combined in differing proportions in
each case, based on the actual symptomatology shown.


In our view the time has passed when one can legitimately
treat schizophrenia by non-physical methods alone, and the
claims that are from time to time made of the necessity of
combining psychotherapy with insulin treatment seem to
be exaggerated. Many patients make an uninterrupted
recovery under insulin treatment without any special psycho-
therapeutic handling whatever. The apparent psychological
precipitants of a schizophrenic illness are frequently found
when the insight gained into the illness is fairly complete to
have been not part causes of the illness but the earliest signs
of its onset. The type of psychotherapy that is most required
with the schizophrenic patient is of a kind that might be found
valuable after any grave physical illness. Once he is re-
covered the patient has to return to the outside world and try
to manage his own affairs without the constant advice of the
doctor; it will often help if the way is made a little smoother
for him. While the disease is in active progress it is hopeless
to try to influence the patient’s ways of thinking; but it is not
so hopeless when the fundamental thought disorder has been
abolished, and there are only a few fragmentary delusional
beliefs, suspicions or foci of apprehension which are left over
as relics from the past.

It is probably beneficial to get into contact with the patient
during the best half hour or so of the day immediately on
wakening after treatment, and to give him then the im-
pression of friendly assistance, even though, as is probable,
any influence one can exert on his morbid ideas is of the most
trifling duration. As soon as he is well enough he should
be kept fully occupied in the afternoon with occupation
therapy, gardening, games, walks, visits to the cinema, and
so on. When recovery has occurred the possibilities of
explanation and reassurance are much more favourable.
The patient will probably have a very clear memory for a
time of his many morbid experiences, and will be anxious to
get some explanation for them. Reassurance will also often
be required on the subject of relapse. It should not be
concealed from the patient that relapse may occur, but he
should be told that the prospects of treatment need be no
worse, should a relapse occur, than they have been with his
first illness, and that if he only seeks advice in the early days
of such a recurrence, they are bright indeed; he should, of
course, on the other hand, be encouraged not to worry about
himself, nor to keep forever a finger on the pulse of his mind.


When recovery has occurred it is advisable to discharge
the patient from hospital as rapidly as possible. Return to
the normal environment, and an abbreviated recollection of
the hospital atmosphere will aid the patient in the recovery of
his self-confidence and powers of adaptation. But it is ad-
visable to keep an eye on the patient in an out-patient clinic
fairly regularly for a month or so, and then it is wise to ask for
three-monthly attendances for another year. If there is the
slightest sign of any relapse it should be promptly dealt with,
and though a relapse may occur at any time up to many
years later, it is most likely to occur fairly soon.

Feedback about this article?

Click on these buttons to visit our journalsSearch for Papers and Articles
Return to Psychiatry On-LineReturn to Dentistry On-LineReturn to Vet On-LineReturn to GP On-LineLink to Pharmacy on-LineReturn to Anaesthesia On-LineReturn to Medicine On-LineReturn to Family Medical Practice On-Line
Chest Medicine
Family Medical
Practice On-Line

All pages copyright ©Priory Lodge Education Ltd 1994-2004.