Postgrad Med J: first published as 10.1136/pgmj.49.571.309 on 1 May 1973. Downloaded from http://pmj.bmj.
com/ on October 29, 2022 at India:BMJ-PG
Postgraduate Medical Journal (May 1973) 49, 309-313.
Stumbling blocks in the study of diamorphine
R. G. TWYCROSS
M.A., B.M., M.R.C.P.
St Christopher's Hospice, Lawrie Park Road, London SE26 6DZ
Summary
Although the centenary of the discovery of diamorphine
is soon to be celebrated, much work remains to be
done before its mode of action is thoroughly under-
stood. The main stumbling blocks facing the would-be
investigator can be summarized as follows:
(1) Diamorphine in solution has a limited shalf-life.
(2) The potency ratio of diamorphine to morphine
is still a matter of dispute.
(3) No basic studies appear to have been carried out
on the fate of orally administered diamorphine.
(4) Current methods of assay are not sensitive
enough to detect diamorphine or its metabolites in
serum after the administration of the drug when given
in therapeutic amounts.
(5) The undesirable side-effects of diamorphine may
vary according to posture and mobility.
(6) As the metabolic handling of the drug may be
different in the two sexes, provision for a between-sex
comparison should be included in the design of a
clinical trial of diamorphine.
Introduction
Diacetylmorphine was first prepared in 1874 by
C. R. A. Wright at St Mary's Hospital, London
(Wright, 1874). Initially it was largely ignored by the
medical profession but some 20 years later interest
rapidly developed as a result of animal experiments
conducted in Germany (Dreser, 1898). Commercial
production was started in 1898 by the Bayer Com-
pany who marketed the alkaloid under the trade
name of 'Heroin'. The name was probably derived
from 'heroisch' which in German medical termino-
logy means large, powerful, extreme, something with
pronounced effect even in small doses. Later this
name became a synonym for the drug.
The new drug received a spontaneous and wide-
spread acceptance comparable with that of drugs
like penicillin and cortisone in more recent years. It
was used initially in a variety of respiratory condi-
tions, such as 'dyspnoea', pharyngitis, laryngitis,
bronchitis, asthma and pulmonary tuberculosis.
Later, hay fever, colds, coughs, pertussis and
pneumonia were added to the advertisers' list of
indications for its use. It was also recommended, in
the United States of America, as a remedy for mor-
phine dependence. Unfortunately this only served
to introduce diamorphine to the addict population
of that country and resulted in increasing abuse.
Its reputation as a potent analgesic developed later
but, by then, its use in so wide a variety of respiratory
conditions was falling into disfavour. Eventually,
because of the growing addiction problem the United
States banned its medical use in 1924. Since then all
but a few countries have fallen into line with what
subsequently became the policy of the League of
Nations and, later, the World Health Organization.
Sponsored. Protected by copyright.
The most recent attempt to bring the United
Kingdom into line with World Health Organization
policy was made in Parliament in 1955. The attempt
failed, but many doctors, especially general practi-
tioners, have felt less able than formerly to use dia-
morphine. Since then a number of studies have been
completed which suggest that in both coronary and
surgical patients no real difference between dia-
morphine and morphine exists (e.g. Scott & Orr,
1969; Morrison et al., 1971). Even so, many still
believe there are real differences in the effects of the
two drugs. For example, physicians working with
terminal cancer patients claim on behalf of dia-
morphine that:
(1) Diamorphine causes less nausea and vomiting
than morphine.
(2) The administration of diamorphine to patients
with anorexia often results in a return of appetite.
This is not the case with morphine.
(3) Diamorphine is less constipating.
(4) Diamorphine enhances the mood of a patient,
whereas morphine not infrequently causes dysphoria.
This does not mean the patient becomes frankly
euphoric; the improvement of mood can be seen as
a return towards normal.
(5) Patients receiving diamorphine tend to be more
alert, active and co-operative compared with patients
receiving morphine.
(6) Diamorphine is a more reliable antitussive
agent.
(7) Diamorphine is more effective at relieving the
anxiety of patients suffering from 'malignant
dyspnoea'.
It is important to realise that this is a list of
supposed benefits. None has definitely been shown
by controlled clinical trial. Nevertheless, it would be

310 R. G.
unwise to disregard without question the consider-
able weight of clinical impression that has accumu-
lated over a period of many years through its use in
this sphere. On the other hand, those antagonistic to
the continuing British use of diamorphine have not
been silent:
'There is a widespread misconception that heroin
has effects significantly different from those of
morphine. It does not, and this misconception should
be dispelled permanently' (Report, 1962).
'Heroin is a drug with no legitimate use. No drug
is more habit-forming; none has more deleterious
effcts' (Washington Post, 1953).
'Heroin should never be used for premedication
since even a single dose can lead to addiction'
(Today's Drugs, 1964).
In summary
'It is clear that almost all comparisons between
morphine and heroin rest upon clinical impressions
rather than experimental evidence. This situation is
unsatisfactory and needs to be put right' (Consumers'
Association, 1967).
However, there are hidden dangers in the clinical
investigation of diamorphine and as these have not
always been appreciated, it is possible that some, if
not most, of the clinical trials completed in recent
years may be of little or no value.
Stability
In any clinical trial, information about the
stability of the preparation being studied is of funda-
mental importance. Morphine appears to be stable
whether stored in crystalline form or in solution.
Unfortunately, such is not the case with diamorphine.
Although crystalline preparations may be stable
(Table 1), in solution diamorphine hydrolyses first
to O6-monacetylmorphine and then to morphine
itself (Table 2). The rate of hydrolysis depends on a
number of factors such as the initial acidity of the
solution, the presence or absence of a suitable
buffering agent and the storage temperature (Davey
& Murray, 1969). It is possible, therefore, that
clinical trials comparing diamorphine and morphine
have, in fact, compared an O6-monacetylmorphine-
morphine mixture with morphine. It should, how-
ever, be recalled that until less than 20 years ago
much of the diamorphine administered in British
hospitals was prepared immediately before injection
TABLE 1. Stability of crystalline diamorphine
hydrochloride (after Lerner & Mills, 1963)
Manufactured Analysis in 1962
In By % Diamorphine % 06-MAM
1926 Bayer 98 2 As many of the terminal cancer patients prescribed
1930 Merck 95 5
Postgrad Med J: first published as 10.1136/pgmj.49.571.309 on 1 May 1973. Downloaded from http://pmj.bmj.com/ on October 29, 2022 at India:BMJ-PG
Twycross
TABLE 2. Serial analysis of a 1% solution of
diamorphine hydrochloride (after Rizzotti, 1935)
Time of
storage Diamorphine 06-MAM Morphine
(months) % % %
0 10 --
3 0-7 0-3
8 0-3 07
12 0-05 09 Trace
16 - 0.9 001
19 - 08 02
22 - 07 0.3
by dissolving a tablet of diamorphine hydrochloride
in water on a teaspoon. It was prior to this date that
the claims for diamorphine's superiority were
originally made and it is since this date that many of
the 'no difference' studies have been completed
(Scott & Orr, 1969; Dundee et al., 1966; Morrison
et al., 1971; Muir et al., unpublished observations).
Of these only Muir and his associates appear to have
been aware of the need to take precautions to
Sponsored. Protected by copyright.
preserve the purity of diamorphine used (Muir,
1972). This they achieved by deep freezing the
solution for injection until just before use.
Potency ratio
A true comparison of the side-effects of any two
therapeutic agents can only be made when equally
effective doses are compared. In most trials in which
the two have been compared it has been assumed
that diamorphine is twice as potent an analgesic as
morphine. Therefore, the drugs have been used in a
1: 2 ratio. That this is an over-simplification can be
seen from work ofBeecher and his associates (Reichle
et al., 1962). They specifically set out to assess the
comparative analgesic potency of the two drugs in
post-operative patients. They concluded that there is
no single dose of diamorphine which is equianalgesic
to 10 mg of morphine. The ratio they obtained varied
between 1: 4 and 1: 2. If a comparative study were
conducted using a 1:4 ratio diamorphine would
almost certainly be shown to have significantly less
unwanted side-effects (Seevers & Pfeiffer, 1936).
In terminal cancer, a between-patient comparison
measuring the side-effects at the effective analgesic
dose would, in theory, be the simplest way to test the
claims of the 'diamorphine is best' school. It would
by-pass the difficult problem of determining the equi-
analgesic dose for each patient. However, in practice,
this approach has several major limitations and it
seems likely that a within-patient cross-over com-
parison will be the only way to settle the present
controversy.
The fate of orally administered diamorphine
diamorphine receive it as an oral mixture the effect

Stumbling blocks in the study of diamorphine
on it of gastric and upper intestinal juices needs to
be considered. Several papers refer to the alimentary
absorption of morphine but little appears to be
known about the fate of orally administered dia-
morphine. It is known that diamorphine is absorbed
fairly well sublingually whereas morphine is not
(Walton, 1935) and that in the normal stomach
virtually no absorption of either drug occurs
(Travell, 1940). Necropsy studies have suggested
that diamorphine is stable in the acid medium of the
stomach (McNally, 1916). However, nothing is
known about the effect of the alkaline upper intestinal
juices on diamorphine. Do they, by facilitating the
rapid absorption of diamorphine, allow it to reach
the circulation intact? Or do they, in view of the
known hydrolysis-accelerating effect of alkalinity,
bring about the rapid degradation of diamorphine
to O6-monacetylmorphine and morphine? These and
other questions remain unanswered.
Furthermore, even if diamorphine can survive the
double hazard of gastric acidity and upper intestinal
alkalinity, can it survive the degrading action of liver
enzymes ? Evidence from animal experiments suggests
not (Wright, 1942). In vivo human studies have yet
to be attempted.
Diamorphine assay
The official pharmacopoeial assay for diamorphine
is colorimetric and requires the preliminary hydro-
lysis of the sample to morphine (British Pharma-
copoeia, 1968). The morphine is then estimated as the
yellow compound nitroso-morphine, formed by
adding sodium nitrite in ammonia buffer. By estimat-
ing the morphine content before and after hydrolysis,
the degree of degradation of a solution of dia-
morphine can be assessed. However, the method has
two major limitations. Although it can be used to
measure the degree of degradation of diamorphine
solutions it is far too insensitive to assess the
presence of either diamorphine or morphine in urine
or serum after medicinal administration. Secondly,
as the nitrosoamine reaction is characteristic of all
phenolic compounds, the method cannot differentiate
between the individual phenolic alkaloids, morphine
and O 6-monacetylmorphine.
In recent years other techniques have been in-
vestigated in an attempt to overcome these two
problems (Gold, 1971). Currently chromatographic
methods are the most widely used (Nakamura &
Ukita, 1967; Davey et al., 1968; Curry & Patterson,
1969), though a quicker and cheaper method for
routine use based on electronic spin resonance is
being used in some centres (Robinson, 1972).
As a result of these advances solutions of dia-
morphine can now be analysed quantitatively with a
high degree of accuracy. However, to measure
diamorphine and its degradation products, 06-
Postgrad Med J: first published as 10.1136/pgmj.49.571.309 on 1 May 1973. Downloaded from http://pmj.bmj.com/ on October 29, 2022 at India:BMJ-PG
311
monacetylmorphine and morphine, in simple solu-
tion is comparatively easy compared with measuring
the same substances in serum. Quantities of morphine
as small as 25 ng have been identified (Wilkinson &
Way, 1969) but in serum and urine the presence of
considerable 'background noise' means that 200 ng/
ml is about the lowest concentration one can reason-
ably hope to detect in practice (Blackmore et al.,
1972). This is not nearly sensitive enough to detect
a drug which is active at concentrations far lower
than this.
With the development of a more sensitive technique
the answers to several questions concerning the fate
of orally administered diamorphine could be obtained
with relative ease.
Mobility
Studies assessing the incidence of vomiting follow-
ing the administration of morphine demonstrate a
clear relationship to mobility. For example, vomiting
occurred in only 2-3 % of 776 patients given morphine
Sponsored. Protected by copyright.
at the Massachusetts General Hospital (Lee, 1942).
These were post-operative patients confined to bed.
On the other hand various physiological studies of
morphine reveal a very high incidence of vomiting:
57% (Ivy et al., 1945) and 93% following 16 mg
(Chapman et al., 1943) and 100% after the injection
of 30 mg morphine (Wolff et al., 1940).
Since it was unlikely that groups of individuals
vary so widely in their response to a drug, a possible
explanation for the widely divergent figures was felt
to lie in the different experimental conditions. It
seemed probable that the divergence could be
accounted for in terms of recumbency and mobility.
This hypothesis was thoroughly tested by Comroe &
Dripps (1948). In one test the incidence of nausea in
a group of postoperative patients up to 2 hr after a
subcutaneous injection of 15 mg of morphine was
12 %. The corresponding figure in normal ambulatory
subjects was 90%. As many of the cancer patients
receiving opiate-containing mixtures orally are not
bedfast, it is not unreasonable to suggest that a real
difference between diamorphine and morphine might
exist when used regularly in such patients whereas
none might be noted in bedfast patients receiving an
opiate on no more than one or two occasions post-
operatively.
Sex difference
It has been suggested that the female rat is able to
deacetylate diamorphine more quickly than the male
rat (Wright, 1942) but whether or not a similar
difference exists between the sexes in man is not
known. However, in the British National Formulary
(1968) it is said that morphine commonly causes
nausea and vomiting, especially in women. It is a
matter of regret, therefore, that either those clinical

312 R. G.
trials known to the author have been limited to one
sex (Smith et al., 1962; Dundee et al., 1966) or the
results have not included a between-sex comparison
(Scott & Orr, 1969; Morrison et al., 1971; Muir et al.,
unpublished observations).
Conclusion
Diamorphine, in some ways, appears to be a
phenomenon rather than just another drug. It is felt
by some to be one of the biggest unnatural disasters
ever to afflict mankind but others continue to regard
it as the unrivalled analgesic for all forms of severe
pain especially that of terminal cancer.
In the interest of science and for the possible
benefit of future patients the need for a scientifically
conducted trial comparing diamorphine and mor-
phine in patients with advanced malignant disease is
apparent. If diamorphine is the better drug for
patients of this kind then the reluctance of doctors
to use it will have to be overcome. If diamorphine is
shown to be no better or, indeed, worse than mor-
phine there would remain little justification for the
Government of the United Kingdom continuing to
sanction its medical use in view of the policy of the
World Health Organization. It should, however, be
stated that the contribution to the total problem of
heroin addiction by diamorphine prescribed on
medical grounds is negligible. Banning the licit
manufacture of diamorphine will do nothing to stem
the flow of illicitly produced material reaching this
or other countries.
References
BLACKMORE, D.J., CURRY, A.S., HAYES, T.S. & RUTTER, E.R.
(1971) Automated analysis of urine for drugs. Clinical
Chemistry, 17, 896.
BRITISH NATIONAL FORMULARY. (1968) p. 70. British Medical
Association and Pharmaceutical Society of Great Britain,
London.
BRITISH PHARMACOPOEIA. (1968) p. 314. Parmaceutical Press,
London.
CHAPMAN, W.P., ARROWGOOD, J.G. & BEECHER, H.K. (1943)
The analgetic effect of low concentrations of nitrous oxide
compared in man with morphine sulphate. Journal of
Clinical Investigation, 22, 871.
COMROE, J.H. & DRIPPS, R.D. (1948) Reactions to morphine
in ambulatory and bed patients. Surgery, Gynaecology and
Obstetrics, 87, 221.
CONSUMERS' ASSOCIATION. (1967) Heroin (diamorphine).
Drug and Therapeutics Bulletin, 5, 11.
CURRY, A.S. & PATTERSON, D.A. (1970) A procedure for the
analysis of illicit diamorphine samples. Journal ofPharmacy
and Pharmacology, 22, 198.
DAVEY, E.A., MURRAY, J.B. & ROGERS, A.R. (1968) The
determination of diamorphine by thin-layer chromato-
graphy and spectrophotometry. Journal of Pharmacy and
Pharmacology, 20, 51s.
DAVEY, E.A. & MURRAY, J.B. (1969) Hydrolysis of dia-
morphine in aqueous solutions. Pharmaceutical Journal,
203, 737.
Postgrad Med J: first published as 10.1136/pgmj.49.571.309 on 1 May 1973. Downloaded from http://pmj.bmj.com/ on October 29, 2022 at India:BMJ-PG
Twycross
DRESER, H. (1898) Parmacologisches fiber einige Morphin-
derivate. Deutsche medizinische Wochenschrift, 24, Vereins
Beilage 185.
DUNDEE, J.W., LOAN, W.B. & CLARKE, R.S.J. (1966) Studies
of drugs given before anaesthesia. X. Diamorphine
(heroin) and morphine. British Journal of Anaesthesia, 38,
610.
GOLD, E.W. (1971) The determination of diamorphine and
cocaine by reflectance spectroscopy. M.Sc. Thesis, Heriot-
Watt University, Edinburgh.
IVY, A.C., GOETZL, F.R. & BURRILL, D.Y. (1944) Morphine-
dextroamphetamine analgesia. War Medicine, Chicago, 6,
67.
LEE, L.E. (1942) Studies of morphine, codeine and their
derivatives. XVI. Clinical studies of morphine, methyl-
dihydromorphinone (Metopon) and dihydrodesoxymor-
phine-d (Desmorphine). Journal of Pharmacology and
Experimental Therapeutics, 75, 161.
LERNER, M. & MILLS, A. (1963) Some modern aspects of
heroin analysis. Bulletin on Narcotics, 15, no. 1, 37.
MCNALLY, W.D. (1916) A report of two cases of fatal heroin
poisoning. Journal of Laboratory and Clinical Medicine, 2,
570.
MORRISON, J.D., LOAN, W.B. & DUNDEE, J.W. (1971) Con-
trolled comparison of the efficacy of fourteen preparations
in the relief of postoperative pain. British Medical Journal,
Sponsored. Protected by copyright.
3, 287.
MUIR, A.L., LAWRIE, D.M., DONALD, K.W. & HARDING,
E.F. A comparison of the circulatory and respiratory
effects of morphine and heroin in patients with acute
myocardial infarction. Unpublished observations.
MUIR, A.L. (1972) Personal communication.
NAKAMURA, G.R. & UKITA, T. (1967) Paper chromatography
study of in vitro and in vivo hydrolysis of heroin in blood.
Journal of Pharmacological Sciences, 56, 294.
REICHLE, C.W., SMITH, G.M., GRAVTNSTEIN, J.S., MACRIS,
S.G. & BEECHER, H.K. (1962) Comparative analgesic
potency of heroin and morphine in postoperative patients.
Journal of Pharmacology and Experimental Therapeutics,
136, 43.
REPORT OF AD HOC PANEL ON DRUG ABUSE. (1962) Confer-
ence on Narcotic and Drug Abuse, Washington, September
1962.
RIZZOTTI, G. (1935) Contributo allo studio delle alterazioni
delle soluzioni acquose di eroina. Archives Internationales
de Pharmacodynamie et de Therapie, 52, 87.
ROBINSON, A.E. (1972) Personal communication.
SCOTT, M.E. & ORR, R. (1969) Effects of diamorphine,
methadone, morphine, and pentazocine in patients with
suspected acute myocardial infarction. Lancet, i, 1065.
SEEVERS, M.H. & PFEIFFER, C.C. (1936) A study of analgesia,
subjective depression and euphoria produced by morphine,
heroin, dilaudid and codeine in the normal human subject.
Journal of Pharmacology and Experimental Therapeutics,
56, 166.
SMITH, G.M., SEMKE, C.W. & BEECHER, H.K. (1962) Sub-
jective effects of heroin and morphine in normal subjects.
Journal of Pharmacology and Experimental Therapeutics,
136, 47.
TODAY'S DRUGS. (1964) Drugs for premedication. British
Medical Journal, 2, 737.
TRAVELL, J. (1940) The influence of the hydrogen ion con-
centration on the absorption of alkaloids from the stomach.
Journal of Pharmacology and Experimental Therapeutics,
69, 21.
WALTON, R.P. (1935) Absorption of drugs through the oral
mucosa. II. Proceedings of Society for Experimental
Biology and Medicine, 32, 1486.

Stumbling blocks in the study of diamorphine
WASHINGTON POST, 4 November 1953.
WILKINSON, G.R. & WAY, L.E. (1969) Sub-microgram
estimation of morphine in biological fluids by gas-liquid
chromatography. Biochemical Pharmacology, 18, 1435.
WOLFF, H.G., HARDY, J.D. & GOODELL, H.J. (1940) Studies
on pain. Measurement of the effect of morphine, codeine,
and other opiates on the pain threshold and an analysis of
Postgrad Med J: first published as 10.1136/pgmj.49.571.309 on 1 May 1973. Downloaded from http://pmj.bmj.com/ on October 29, 2022 at India:BMJ-PG
313
their relation to the pain experience. Journal of Clinical
Investigation, 19, 659.
WRIGHT, C.I. (1942) The deacetylation of heroin and related
compounds by mammalian tissues. Journal of Pharmacology
and Experimental Therapeutics, 75, 328.
WRIGHT, C.R.A. (1874) On the action of organic acids and
their anhydrides on the natural alkaloids. Journal of the
Chemical Society, 12, 1031.
Sponsored. Protected by copyright.