Klin Wochenschr (1982) 60: 896-913
Klinische
Wochen-
schrift
© Springer-Verlag 1982
Definition and Classification
of the Histamine-Release Response to Drugs
in Anaesthesia and Surgery:
Studies in the Conscious Human Subject* **
W. Lorenz 1, A. Doenicke 2, B. Schrning 3, Ch. Ohmann 1, B. Grote 2, and E. Neugebauer 1
1 Division of Theoretical Surgery, Department of Operative Medicine I, University of Marburg, Robert-Koch-Str. 8, D-3550 Marburg,
Federal Republic of Germany
2 Division of Anaesthesia, Surgical OPC Clinic, University of Munich, D-8000 Munich, Federal Republic of Germany
3 Division of Anaesthesia, Orthopedic Clinic, University of Heidelberg, D-6900 Heidelberg, Federal Republic of Germany
Summary. In 2 clinical studies in 40 conscious human volun-
teers and 164 orthopedic patients histamine-release re-
sponses were diagnosed, defined and classified. Polygeline
(Haemaccel) in its now outdated formulation [40] was cho-
sen as a clinical histamine releaser. The main interest was
not concentrated on the extreme, the "classical" anaphy-
lactic response, but on the
average
histamine-release re-
sponse found in clinical experiments with so many drugs
in the last 10 years.
In human volunteers 600 ng/kg histamine was i.v. in-
jected. Indicants for a systemic anaphylactoid reaction with
the highest incidence ratio were tachycardia, plasma hista-
mine levels > t ng/ml, "metallic taste", flush, congestion
of head, "wet eyes" and tears, hypertension and headache.
Following polygeline none of these subjects developed a
life-threatening reaction, but 12 showed a systemic re-
sponse, 11 a cutaneous reaction and 17 were non-re-
sponders. Indieants for a systemic anaphylactoid reaction
with the highest incidence ratio were plasma histamine lev-
els > 1 ng/ml, tachycardia, wheals, sensation of heat, nar-
rowness of throat, hypertension, headache and wet eyes
or tears.
In a prolective, cohort study in the orthopedic patients
3 subjects with life-threatening reactions, 27 with systemic
response, 96 with cutaneous reaction and 38 non-re-
sponders were included. Indicants with the highest inci-
dence ratio were tachycardia, plasma histamine levels
> 1 ng/ml, erythema and wheals, cough, flush, stuffy nose
and facial oedema. With this trial the indicants for diagnos-
ing a systemic histamine release response in volunteers were
validated in patients to a large extent.
Thus the average histamine-release response was defined
by clinical signs such as tachycardia and mild hypertension,
scattered hives such as spots of erythema and wheals, respi-
ratory symptoms in the laryngeal and nasal region, such
as cough, narrowness in the throat, stuff), nose and sneezing
and
by pathological plasma histamine levels (> 1 ng/ml), in
addition histamine-release responses were differentiated as
cutaneous responses, systemic responses and life-threaten-
ing responses by clinical and operational criteria and by
plasma histamine levels. Using clinical trials and medical
* Dedicated to Professor Otto Lindenschmidtt
** Supported by grant of Deutsche Forschungsgemeinschafl (Lo
199/10 and Lo 199/13-6)
Offprint requests to:
Professor Dr. W. Lorenz (address see above)
decision making procedures the incidence of systemic hista-
mine-release responses in patients higher by two orders of
magnitude than in other studies reported hitherto.
Key words: Histamine release - Diagnosis -Volunteers -
Patients - Medical decision making.
Definition und Klassifikation yon Histaminfreisetzung
nach Gabe von Arzneimitteln in An~isthesie
und Chirurgie: Studien am wachen Menschen
Zusammenfassung:: In 2 klinischen Studien bei~ 40 wachen
Freiwiltigen und 164 orthop/idischen Patienten wurde ver-
sucht, Histaminfreisetzungsreaktionen zu diagnostizieren,
zu definieren und zu klassifizieren. Haemaccel in einer heute
klinisch nicht mehr verwendeten Zubereitung [40] wurde
als klinischer Histaminfreisetzer verwendet. Das Hauptinte-
resse gait nicht der extremen, der klassischen anaphylak-
tischen Reaktion, sondern einer durchschnittlichen Hist-
aminfreisetzung, die in ktinischen Untersuchungen der letz-
ten 10 Jahre mit so vielen Arzneimitteln gefunden wurde.
Bei den Freiwitligen wurden 600 ng/kg Histamin intra-
venrs verabreicht. Indikatoren ffir eine systemische anaphy-
laktoide Reaktion mit der hrchsten Inzidenzrate waren Ta-
chykardie, Plasmahistaminspiegel fiber 1 ng/ml, metal-
lischer Geschmack, Flush, Kopfdruck, feuchte Augen oder
Trfinen, Hypertension und Kopfschmerzen. Nach Haemac-
cel-Infusion zeigte keiner der Probanden eine lebensbedroh-
liche Reaktion, abet 12 eine systemische und 11 eine Haut-
reakfion, w/ihrend bei 17 keine Symptome gefunden werden
konnten. [ndikatoren mit der hrchsten Inzidenzrate waren
wiederum Plasmahistaminspiegel fiber 1 ng/ml, Tachykar-
die, Quaddeln, Hitzegeffihl, Enge im Hals, Hypertension,
Kopfschmerzen und Tr/inen.
In einer prolektiven Cohortstudie wurden aus 600 or-
thop~idischen Patienten t64 ausgew~ihlt: 3 hatten eine le-
bensbedrohliche Reaktion, 27 eine systemische und 96 eine
Hautreaktion, 38 Patienten zeigtcn keine Symptome. tndi-
katoren mit der hrchsten Inzidenzrate waren wiederum Ta-
chykardie, Plasmahistaminspiegel fiber 1 ng/ml, Erytheme
und Quaddeln, Husten, Flush, verstopfte Nase und Ge-
sichtsrdem. Damit wurden dutch die Patientenstudie die
Indikatoren f/it eine systemische Histaminfreisetzungsreak-
tion in Probanden zu einem grol3en Teil validiert. So 1/il3t
sich eine durchschnittliche Histaminfreisetzungsreaktion als
eine systemische anaphylaktoide Reaktion charakterisieren,
mit klinischen Symptomen wie Tachykardie und leichte Hy-
pertension, verstreuten Effloreszenzen, respiratorischen
Symptomen im Bereich des Kehlkopfs und der Nasen-
schleimhaut und durch pathologische Plasmahistaminspie-
gel (>1 ng/ml). Augerdem wurden die Histaminfreiset-
zungsreaktionen in kutane, systemische und lebensbedroh-
liche Reaktionen eingeteilt, wobei klinische und operafio-
nale Kriterien sowie Plasmahistaminspiegel fiir die Klassifi-
kation verwendet wurden.
Sehliisselwiirter: Histaminfreisetzung- Diagnose- Proban-
den - Patienten - Medizinische Entscheidungsfindung.
Introduction
Before reliable and practicable [20] histamine assays in hu-
man plasma became available for routine use [4, 26, 27]
a histamine-release response to drugs could not be esti-
mated in clinical conditions with an acceptable accuracy
[9, 10, 28, 29, 34, 36, 45, 51, 58, 62]. But even one decade
after the discovery of H2-receptor antagonists [5] and the
development of these highly sensitive and specific asays [4,
26] histamine was still not generally recognized as an impor-
tant mediator of any pathological and clinical phenomenon
[63]. It was something of a paradigm [22a] not to assign
any distinct function to this biogenie amine - neither qualita-
tively nor quantitatively [13].
In the labyrinth of intricate views on histamine this dis-
appointing situation could be amended by only step by
step. A very early, necessary and urgent one was to estab-
lish the diagnosis of the histamine-release response to an
individual drug in clinical conditions, especially in anaesthe-
sia and surgery where so many histamine-releasing drugs
are routinely administered intravenously [39]. The main in-
terest in this study therefore was not concentrated on the
extreme, the "classical anaphylactic" response, but on the
average response which was found in clinical experiments
with so many drugs during the last 10 years. Thus two
aims were devised in this investigation:
(l) In conscious human volunteers clinical symptoms
and plasma histamine levels were assessed after intravenous
injection of histamine. A dose was chosen which increased
the plasma histamine levels to such an extent as the average
histamine-release response to drugs used in anaesthesia [28,
34]. In addition, the same subjects received one of the "clin-
ical" histamine releasers, polygeline (Haemaccel) [29, 34,
39, 40) a short time after exogenous histamine. From all
the findings a questionnaire with many items was developed
to diagnose an average histamine-release response.
(2) Using these items in conscious orthopedic patients
the clinical symptoms, biological reactions and plasma his-
tamine levels were assessed which were elicited by Haemac-
eel, the same histamine liberator as used in the volunteers.
By this study the set of indicants for diagnosing a histamine-
release response was validated. Parts of this very extended
study have already been published in other journals [1,
35-40]. The subject of this presentation, however, was not
considered in the previous communications.
Materials and Methods
1. Theoretical and Ethical Issues [41]
The aim of the study could not be achieved by a single technique
(e. g. a single biochemical test). Among several approaches a de-
897
(1)
Was the drug really given or taken?
(2)
Symptoms similar to those elicited
~
by free histamine?
(3) ~ ~ (3) Changes in
histamine content of
-/ \+
-/ ~4
tissues and body fluids?
[~](4)~ [~] ~ (4)
Exogenous histamine
/ ~ // ~ as contamination ?
(5) ~ .C~ (5) No other drugs given at the
-/ "~ #f ~ same time?
/
"/ \+ / -/ \÷
mine releasers?
~7) ~ ~ el ~ ~ ~ (7) oo~s the ~e~
~-
~ ~/ ~] ~ ~// ~
lease h|stamme'~
subclinical clinical
OK! The patient has a histamine-release response to the drug
Fig. 1, Decision tree for diagnosis of hkstamine-retease response
to an individual drug in an individual patient. The open circles
represent test nodes, the plus or minus branches the preceding
facts or findings which are necessary to establish finally the diagno-
sis. The tree structure was performed according to Lusted [43].
The small squares denote steps of the decision-making process
(dead ends), not choice nodes as in a modified system of displaying
the decision problem [68]. The arrows indicate the final outcome
of a clinical or subclinical histamine-release response
cision-making process was chosen for a first approximation which
was illustrated by a decision tree (Fig. 1). The designs of the two
prospective studies in volunteers and patients payed regard to all
the questions in this structure.
The decision tree showed 4 paths and 3 outcomes comprising
clinical, subclinical and indeterminable histamine-release response
to an individual drug. Subclinical reactions were included since
they were of considerable interest for a second drug exposure and
other complications after surgery [66]. For the aims of the two
clinical trials, however, the second path from the right was the most
important one including symptoms, changes in plasma histamine
content, no contamination with exogenous histamine, no other his-
tamine releasers given and a highly inert solvent system. Since
the final outcome (diagnosis of a clinical response) should depend
mainly on findings obtained by the questions 2 and 3, the studies
had to be conceived in such a way that the other information
was obtained with the highest possible precision and accuracy.
To achieve this aim the following conditions numbered in agree-
ment with the decision tree were chosen:
(1) Exogenous histamine and the histamine releaser polygeline
(Haemaccel) were given to the human subjects by a single investiga-
tor (author), to volunteers by B.G. and to patients by B.Sch.
(2) Clinical signs were recorded by a single observer, too, in volun-
teers by W.L. and in patients by B,Sch. Thus observer variation
within the trials could be avoided. Between the trials it was reduced
by discussions in detail of all defintions of the various indicants
by the observer in volunteers with the observer in patients [16,
43].
In volunteers a dose of 600 ng/kg i.v. exogenous histamine was
expected to produce plasma histamine levels similar to those of
an average histamine-release response (pilot study in volunteers
and in dogs [30]). An injection was chosen because of pharmacokin-
etics more similar to histamine release than histamine infusion.
The intravenous route corresponded to histamine release from the
skin which is common for many histamine liberators [47] including
Haemaccel [30], To avoid tachyphylaxis and prolonged elevation
of plasma histamine content exogenous histamine was applied be-
.fore potygeline.
Polygeline was tested in a dose of 500 ml/3 rain in normovo-
laemic haemodiluation [34] since this design repeatedly has pro-
duced average histamine-release responses [28, 33-35, 54]. Hae-
898
maccel as a plasma substitute was pure and chemically defined
and could be administered in clinical conditions where no other
drug, solvent or treatment interferes [34]. Histamine release by
this drug was predominant [13, 35] i.e. no other mediator seemed
to be of any importance in the reaction to polygeline [35]. Its
incidence was sufficiently high in clinical situations and its mecha-
nisms were reasonably uniform [39, 40, 56].
In patients the infusion of 500 ml/10 min Haemaccel before op-
eration was indicated for therapeutic reasons [62a]. This decision
was made by clinicians responsible for the patients and not by
those involved in the trial. The administration of polygeline in
conscious patients, however, was utilized for this study since anaes-
thesia was considered to modify, but not in general to reduce the
extent of the clinical reaction [46].
(3) All plasma histamine assays were performed by a single techni-
cian who was not aware of the clinical reactions and received the
plasma samples in a randomized sequence. The assays run under
quality control using a mixed control plasma prepared from the
patients of the trial in Heidelberg and control charts according
to Levey and Jennings [24].
(4) Polygeline was tested in several batches to obtain a random
distribution of histamine-release responses over the manufacturing
process. In none of the 4 batches administered to volunteers and
patients could any histamine contamination be detected. In addi-
tion, the fluorescent material in polygeline [31] was shown not
to interfere with the histamine assay using several tests of identifica-
tion [32].
(5) Interference with other drugs was prevented by controlling meti-
culously the time before and after histamine and polygeline admin-
istration. Histamine release was considered as a sudden event with
an explosive velocity of increment for plasma histamine levels [36]
followed by a rather quick elimination (plasma t 1/z for i. v. injected
histamine about 2 min, for histamine release about 5--10 min [30,
361.
Thus in volunteers all subjects were systematically asked for
any drug intake at the day of experiment by a single observer
(W.L.) and 2 of them were excluded since this was the case. Follow-
ing i.v. histamine injection 20 rain elapsed before polygeline was
infused. Basal plasma histamine levels were again attained at this
time (Fig. 5).
In patients only elective surgery was accepted. The night before
operation they received standard treatment (only nitrazepam and
heptabarbitone). In the morning the premedication tbr anaesthe-
sia started with the infusion of polygeline. Inovar (in Germany
known as Thalamonal) and atropine, the other two components
of the premedication were administered not prior than 30 min after
the start of infusion which lasted not longer than 10 rain. Patients
who needed additional drugs for treatment of concomitant diseases
such as/~-adrenoceptor blocking drugs or digitalis received these
compounds very early in the morning and not later than 2 h before
starting the infusion of polygeline.
(6) In clinical conditions none of these drugs was hitherto suspected
to be a histamine releaser [36, 39].
(7) Histamine was dissolved in saline and polygeline in a slightly
modified Ringer solution which was shown not to release histamine
[34].
Ethical Just~cation. In volunteers informed consent was obtained
in written form following an extensive discussion about the impor-
tance and necessity of the study, the experimental design and the
risks of a histamine injection, blood donation and infusion of a
histamine releaser. It was explained that at the time of the study
(March 1976) no colloidal plasma substitute commercially avail-
able was entirely without risk of inducing an anaphylactoid re-
sponse [34, 52, 54]. At the time of the study polygeline it its now
outdated formulation ("classical polygeline" [40]) had been given
to patients all over the world for more than 10 years. Utmost
safety was provided for each volunteer during the clinical experi-
ment by at least 2 anaesthesists being present in the room all the
Table
1. Attributes of the subjects in the volunteer study
Attributes ~ (interquartile
range, range)
Age [years]
Weight [kg]
Sex (male/female)
Profession (S-M, S-NM, HS, NHS)
Allergy in past (yes/no)
Allergy at present (yes/no)
Previous infusion of colloidal
plasma substitutes (yes/no)
26 (23-29, 21-37)
70 (62-75, 50-94)
Ratio of numbers
32/8
12ii3/10/s
9/31
10/30
10/30
S-M student of medicine, S-NM student not studying medicine,
HS hospital staff, NtIS other professions, but not hospital staff
(e. g. civil servant, engineer, photographer). Allergy in past = more
than 9 months ago, allergy at present=from 9 months in the past
until now. Previous infusion of plasma substitutes included 9 sub-
jects who were included in previous trials (the last 2 years before)
time. All emergency equipment was in the room and ready for
use. In addition, all volunteers were asked to inform the investiga-
tors the day after the experiment about their state of health. There
was no complaint by any of them.
In orthopaedic patients no safer alternative treatment was avail-
able. The patients were informed about additional taking of blood
samples for plasma histamine assays. Utmost safety was provided
for each patient in the same way as for each subject in the volunteer
study.
2. Time Schedule, Volunteers and Patients, Materials
The volunteer study was conducted in 40 subjects (Table 1) in Mun-
ich at March 8-13, 1976. For this trial histamine dihydrochloride
puriss. (Fluka, Buchs) was dissolved in sterile saline (2 mg/100 ml,
University's pharmacy). Only 2.5-5 ml were injected to achieve
a reliable bolus injection. - Polygeline (Haemaccel, Behringwerke
Marburg) was infused in 4 batches (Op 3000 and 3812, V-235
and V-244) which were characterized in previous communications
[31, 32, 34, 40] regarding their chemical and pharmacological prop-
erties. The 40 volunteers were assigned to 4 treatment groups (1
batch for one group) by a balanced random allocation with random
digits. This part of the study was a double-blind trial [40]. The
patient study was conducted in 164 subjects in Heidelberg from
March 14, 1977 till January 12, 1978. They were selected from 600
patients in a controlled clinical trial [36, 38-40] who after prolective
stratification comprised both sexes divided into 5 classes of age
(20->60 years, 30 patients in each cell). The selection became
necessary since the plasma histamine assays were too time-consum-
ing and expensive to be carried out in all patients. The procedure
for selection is shown in Table 2: (1) In all participants of the
trial with clinical symptoms of an anaphylactoid reaction [35]
blood was taken for plasma histamine assays. From a previous
trial [54] 30 incidents were expected and fortunately also found.
(2) To avoid seasonal variations or bias introduced by new drugs
or treatment during the study the control group with "no adverse
response" was formed by selecting always the next two patients
following the last one with an anaphylactoid reaction. It was ex-
pected that some of these patients would develop an allergoid reac-
tion (response restricted to the skin) [35]. By that the sample of
true non-responders would be reduced to approximately the same
size as that of the sample with anaphytactoid reactions. In this
aspect the selection was rather successful (Table 2). (3) Then a
second control group was formed with patients "showing only
allergoid reactions" to discriminate between anaphylactoid (sys-
temic) and allergoid (cutaneous) reactions to polygeline. For this
clinically very important classification a compromise was chosen
899
Table 2. Patients in the clinical trial on histamine release by polyge-
line and proportions of those involved in the plasma histamine
assay
Reaction Expected a Found
to polygeline
Patients Pro- Patients Pro-
in trial portion in trial portion
for for
In] [%]b hista- [nl [%] hista-
mine mine
Anaphylactoid 30 5 30 30 5 30
No response 420 70 50 413 69 38
Allergoid 150 25 100 157 26 96
Total 600 100 180 600 100 164
" Frequencies and rate according to Schtning and Koch [54]
b Percentage of patients in the trial. 4 patients in the third group
were lost by failures of centrifugation, 4 were excluded because
they could not unequivocally be classified as "no responders" or
"patients with allergoid reaction". Sex distribution (male/female)
in the three groups: Anaphylactoid 24/6, no response 16/22, aller-
gold 56/40 patients. Age distribution in the three groups (20-29,
30-39, 4049, 50-59, >60 years): Anaphylactoid 6/10/5/6/3 - no
response 7/8/7/8/8 - allergoid 18/20/29/14/15. The differences were
not considered as influencing the incidence ratio of pathological
plasma histamine levels and clinical symptoms
between maximum information and feasibility selecting two-thirds
of patients with allergoid reactions for plasma histamine assays
by random allocation with random digits. Selection and randomi-
zation procedure could only be realized because plasma was pre-
pared from all 600 patients in the study before polygeline infusion.
Plasma of patients which was not used for histamine assays was
pooled for quality control samples.
Polygeline was infused in 4 batches (Op 3939 and 3946, V-244
and V-265). The 600 patients were assigned to 4 treatment groups
(1 batch for one group). This part of the study again was a double-
blind trial. However, no differences were found in the incidence
of reactions between the 4 batches [38].
Reagents for plasma histamine assays were the same as de-
scribed by Lorenz et al. [27, 32].
3. Experimental Design
(1) In volunteers the investigation started with a half-structural
interview including questions on case history to identify special
risks of anaphylactoid reactions to plasma substitutes [34], ques-
tions on previous exposure to intravenous agents [28, 65, 66] and
on clinical signs appearing even before injection of histamine. The
clinical symptoms were chosen from previous reports [23, 59, 69]
and earlier studies [28, 33, 34]. The volunteer was asked to give
his informed consent.
Thereafter in the laying subject leads for the ECG were at-
tached to the extremities and a sphygmomanometer was applied
to the right upper arm. Two Braunulae were inserted in this arm,
one in the cubital fossa for injection of histamine and one more
peripheral for the venous blood collection. A first 20 ml sample
was taken to produce plasma for a pool which was used for quality
control samples. Five minutes later a second sample was taken
to obtain the pre-injection plasma histamine level. Then histamine
was administered and blood was taken at 1, 2, 3, 5 and 10 min
after its application. During this time the volunteer was asked
about and carefully looked for any clinical symptoms. First of
all, spontaneous reactions were recorded, then all questions asked
before the injection were repeated 3 and 10 min after histamine
injection.
Fifteen minutes after histamine injection blood donation
(360 ml) was started from the left arm and 1 min later polygeline
was infused (500 ml, in about 3 min). Blood samples for plasma
histamine were taken 1, 5, 10, 15 and 20 min after the end of
infusion and during this time the clinical symptoms were recorded
as described before. Thoughout the experiment the heart rate was
continuously monitored (lead II in ECG) and the blood pressure
was measured every min.
From the questions, observations and responses of the volun-
teers a questionnaire was developed with indicants for diagnosing
a histamine-release response (Appendix 1).
(2) Also inpatients the investigation started with an interview using
the questionnaire in a protocol ready for electron data processing
[57]. Then pulse rate (by hand) and blood pressure (by sphygmo-
manometer) were measured and a large arm vein was cannulated
by a Braunula No. 1. For plasma histamine assay blood samptes
were taken 5 rain later before polygeline (500 ml) was infused in
10-15 min, at the end of the infusion and again 5 min later to
get parts of a clearance curve as in the volunteers [39]. At the
end of the infusion, 5, 15 and 30 rain later pulse and blood pressure
were measured and any clinical symptoms - especially any skin
reactions - were systematically looked for even by turning the
patient round and inspecting all parts of the body. The protocol
was tested for its completeness and any observations to be added
were specified. For plasma preparation [27] blood was centrifuged
in a Minifuge (Christ, Osterode) in the operation theatre [27].
4. Me~o~
Clinical symptoms and reactions were assessed using the question-
naire described in this communication. - Plasma was prepared
and plasma histamine concentration was determined according to
Lorenz et at. [27] using a sensitive and specific fluorometric-ftuor-
oenzymatic assay. - In all cases with anaphylactoid reactions hista-
mine was identified by fluorescence spectra, :heating test and spec-
tra after the heating test as described by Lorenz et al. [32] and
Parkin et al. [48]. - All assays were performed under quality con-
trol using two plasma pools obtained from volunteers and patients
to which authentic histamine was added to give "plasma" hista-
mine levels similar to those of an average histamine-release re-
sponse (about 3 ng/ml).
5. Statistics and Definitions
For descriptive statistics median, interquartile range, range and
incidence ratio were used, for inferential statistics 95% confidence
intervals and decision-making matrix [15, 43, 68]. The correlation
between plasma histamine levels and increase in heart rate (tachy-
cardia) was analysed by the method of least squares using a Hew-
lett-Packard computer HP 9815 A.
A crucial question in this study was the definition of the "dis-
ease". In the first part of the volunteer study the disease was an
anaphylactoid-like reaction to the i.v. injection of histamine, in
the second part of this study and in the clinical trial the disease
was a true anaphylactoid or allergoid (cutaneous anaphylactoid)
reaction to polygetine. The similarity of the two "diseases" in
terms of diagnosis was one of the most important questions of
this article.
Thus anaphylactoid and allergoid reactions have to be defined
[39]: (1) An anaphylactoid reaction is a clinical reaction with gener-
alized urticaria (more than 5 wheals in different regions of the
body), discomfort (nasal catarrh, narrowness of the throat, ble-
pharoedema, nausea or vomiting, diarrhoea), bronchospasm, ta-
chycardia, hypotension, circulatory and cardiac insufficiency and
arrest. At least generalized urticaria plus discomfort (except dermal
pruritus) were necessary to classify an adverse reaction to drugs
as anaphylactoid. Generalized urticaria plus tachycardia, but also
bronchospasm, cardiac arrhythmia, hypotension or cardiac arrest
with or without urticaria are other combinations of clinical signs
being classified as anaphylactoid. (2) An allergoid (cutaneous ana-
phylactoid) reaction is again a clinical reaction classified by eryth-
ema, urticaria and dermal pruritus only. Lateron in this article
900
A
3.0-
"Normal" type a)
/ \ n = 33
2.0
E
0 ,
II
.=. l
g
i'o I~ 2'o
.2
"Delayed" type b)
tHistamine / ~ n-4
O
c
*=
"E
No 3
tO
J¢
i
o.-- 0 ' III g 10
1'5 20
2"01 ~ "Redistribution" type c)
n--3
1.0 j/
0 P'l III 5 ' 10 1,~ gO
T,me [m,n ]
Fig. 2. Histamine elimination curves following i.v. injection of ex-
ogenous histamine (600 ng/kg). Single values obtained from the
volunteers, n refers to the number of subjects showing the attribute,
No
refers to the number of a single person in the sequence of
the experiments. Since in one of the subjects only the first sample
after histamine injection was lost, but all other values corresponded
to those of the type-(a)-elimination curve this subject was not ex-
cluded in the figure. For further conditions see Materials and
Methods
anaphylactoid reaction will be replaced by systemic histamine-re-
lease response. In terms of taxonomy of diseases the systemic hista-
n-tine-release response is a special case of an anaphylactoid reaction
which in its turn is a special case of a pseudo-allergic reaction
[12, 22, 391.
Another crucial question in this study was the defintion of
a pathological plasma histamine level as indicant for an anaphylac-
toid reaction. (1) It is defined as
top
level obtained by measuring
several values in a Bateman function. (2) It is defined as positive
or pathological if it is higher than 1 ng/ml which is the upper
2 SD limit of the "normal range". Considering well the criticism
of this concept [18] it is emphasized that since 1975 the normal
range of plasma histamine values has been always 0--1 ng/ml in
studies with 120 [33] and 40 [37] volunteers and in 50 [39] and
299 [55] patients of both sexes and 5 classes of age using our
fluorometric-fluoroenzymatic assay [27, 39].
Results
1. Plasma Histamine Levels and Clinical Signs
in Volunteers Following Administration
of Exogenous Histamine
Histamine elimination curves showed 3 typical patterns
(Fig. 2): A "normal" type in 80% of the subjects showed
first-order kinetics, but in 20% a "delayed" type or a "red-
istribution" type was found, probably due to difficulties
with the bolus injection. Such "failures"occur also in clini-
cal routine [39]. For assessing the incidence ratio, however,
the pathological plasma histamine values the one subject
in whom the first plasma was lost and the three with "red-
istribution" type curves were excluded (Fig. 3, Table 3).
From the data in the other 36 volunteers, however, it
was evident, that the imitation of an anaphylactoid reaction
by injection of histamine was reasonably successful (Fig. 3).
Eighty percent of the plasma histamine levels exceeded 1 ng/
ml and the average concentration was about 1.52 ng/ml,
similar to that found in histamine-release responses to an-
aesthetics and plasma substitutes [10, 34]. In 2 subjects plas-
ma histamine levels of about 4 ng/ml were measured, ac-
companied by considerable tachycardia. In 8 subjects in-
creases of plasma histamine levels could be determined, but
were rather small and did not exceed the normal range.
The variation in plasma histamine levels following the same
E
i i
3
8
E
0 5
•
level before histamine injectkm 0.3 (0-0.65)
[]
increase after injection
(6OOng/kg i.v.) 1.0 (0.4-3.5)
(~ (range)) (ng/ml)
10 15 20 25
Subjects with "
anaphylactoid " reaction
[
n
]
m
i
ll, lll,
30 35
Fig. 3. Plasma histamine levels before
and after injection of histamine in
volunteers in whom an anaphylactoid
reaction was imitated by i.v.
administration of only 600 ng/kg
histamine. Single values obtained from
each of the volunteers before and after
histamine injection. The vertical dotted
line indicates the cut-off point for an
anaphylactoid reaction as described in
Methods