British Journal of Anaesthesia 1994; 73: 484-^189
Interactions between mivacurium and atracurium
M. NAGUIB, M. ABDULATIF, A. AL-GHAMDI, M. SELIM, M. SERAJ, M. EL-SANBARY AND
M. A. MAGBOUL
Summary
We have studied the interaction between atracurium
and mivacurium. The dose-response relationships
of atracurium, mivacurium and their combination
were studied in 96 ASA I or II patients during
thiopentone-fentanyl-nitrous oxide-isoflurane
(1.2%end-tidal) anaesthesia. Neuromuscular block
was recorded as the evoked thenar mechano-
myographic response to train-of-four stimulation of
the ulnar nerve (2 Hz at 12-s intervals). The
dose-response curves were determined by probit
analysis. Isobolographic and algebraic (fractional)
analyses were used to assess quantitatively the
combined effect of equipotent doses of atracurium
and mivacurium and to define the type of interaction
between these drugs. Isobolograms were con-
structed by plotting single drug ED.*, points on the
dose co-ordinates and a combined EDso point in the
dose field. The calculated doses producing 50%
depression (EDBO) of the first twitch height were
50.5 (95% confidence intervals 48.9-52.1) and
20.8 (20.3-21.3) ng kg- 1 for the atracurium and
mivacurium groups, respectively. Isobolographic
and fractional analyses of the atracurium-
mivacurium combination demonstrated zero in-
teraction (additivism). An additional 26 patients
anaesthetized with thiopentone-fentanyl-nitrous
oxide-isoflurane were allocated randomly to receive
either atracurium 0.5 mg kg~1 (n = 13) or miva-
curium 0.15 mg kg" 1 (n = 13). Additional main-
tenance doses of mivacurium 0.1 mg kg" 1 were
administered to patients in both groups, whenever
the first twitch recovered to 10% of control. The
duration of the first maintenance dose of miva-
curium to 10% recovery of the first twitch was
greater (P < 0.0005) after atracurium (25 (21.8-
28.5) min) than after mivacurium (14.2 (11.9-
16.6) min). However, the duration of the second
maintenance dose of mivacurium after atracurium
(18.3 (12.6-24) min) was similar to that of miva-
curium after mivacurium (14.6 (10.6-18.6) min).
We conclude that the combination of atracurium
and mivacurium is additive and that the use of
mivacurium after atracurium-induced neuromuscu-
lar block results in increased duration of the first
(but not the subsequent) maintenance dose of
mivacurium. {Br. J. Anaesth. 1994; 73: 484-489)
Key words
Neuromuscular block, atracurium. Neuromuscular block, miva-
curium. Potency, anaesthetic, EDgo
Mivacurium chloride is a non-depolarizing neuro-
muscular blocking agent which has recently been
introduced into clinical practice. Although chemi-
cally related to atracurium, it has a shorter duration
of action [1,2]. This difference in pharmacodynamic
characteristics might present some advantages in
combining the two drugs in clinical practice. One
example might be the necessity for adequate paralysis
during peritoneal closure. A supplementary dose of
atracurium given at this time may be ineffective if the
dose is small; a larger dose may prolong the duration
of action. In addition, the use of suxamethonium to
prolong a waning non-depolarizing block is not
advisable as a block of unpredictable nature and
duration may result [3,4]. It is not known if the use
of mivacurium in the presence of residual atracurium
activity enables the anaesthetist to adequately control
the depth and duration of neuromuscular block.
It has been shown that concomitant administration
of some mixtures of non-depolarizing neuromuscular
blocking compounds may result in additive [5-7] or
synergistic effects [5,8,9]. Combinations of atra-
curium and mivacurium have not been studied
previously. This study was undertaken to charac-
terize the interaction of a combination of atracurium
and mivacurium in humans using the isobolographic
method. This method makes no assumption re-
garding mechanism of action or shapes of the
dose-response relationships [10,11].
Patients and methods
After obtaining local Ethics Committee approval and
informed consent, we studied 122 ASA I or II
patients of both sexes, aged 18-55 (mean31.1)yrand
weights 45-85 (mean 67.3 (SD 11.2)) kg. All patients
were undergoing elective procedures, had no neuro-
muscular, hepatic or renal disease, and were not
receiving any drugs known or suspected of inter-
fering with neuromuscular function. All patients
MOHAMED NAGUIB, MB, BCH, MSC, FFARCSI, MD, MOHAMED
ABDULAT1F*, MB, BCH, MSC, MD, ABDULMOHSIN A L - G H A M D I * , MB,
BS, DA, MAHDI SELIM*, MB, BCH, MSC, MOHAMED SERAJ, MB, BCH,
DA, FFARCSI, MOHGA E L - S A N B A R Y * , MB, BCH, MSC, MD, MAGBOUL
A. MAGBOUL, MB, BS, FFARCSI, Department of Anaesthesia and
ICU, King Saud University, Faculty of Medicine at King Khalid
University Hospital, P.O. Box 7805, Riyadh 11472, Saudi Arabia.
Accepted for publication: April 15, 1994.
* Present address: Department of Anaesthesia, King Faisal
University, Saudi Arabia.
Atracurium-mivacurium interactions
were premedicated with lorazepam 2 mg orally
approximately 90 min before induction of anaes-
thesia. An i.v. infusion of lactated Ringer's solution
was commenced before induction of anaesthesia.
ECG, pulse oximetry and arterial pressure were
monitored. Temperature was monitored by a
nasopharyngeal thermistor and maintained at
36.5±0.5 °C.
INTERACTION STUDIES
Anaesthesia was induced with thiopentone 5 mg kg"1
and maintained with 60% nitrous oxide and iso-
flurane in oxygen via a face mask and supplemented
with fentanyl 2ugkg~'. The trachea was sprayed
with 4 % lignocaine 4 ml and was intubated without
the use of neuromuscular blocking agents. After
intubation, the end-tidal concentration of isoflurane
was adjusted to 1.2% (1 MAC excluding nitrous
oxide). The concentrations of isoflurane, nitrous
oxide, oxygen and carbon dioxide were measured
continuously by a multiple-gas analyser (Capnomac,
Datex Instrumentarium Corporation, Helsinki,
Finland). Ventilation was adjusted to maintain
normocapnia (end-tidal carbon dioxide pressure
4.8-5.3 kPa).
The ulnar nerve was stimulated supramaximally at
the wrist with square pulses of 0.2-ms duration,
delivered in a train-of-four (TOF) sequence at 2 Hz
every 12 s, using a Myotest peripheral nerve stimu-
lator (Biometer International, Odense, Denmark).
The resultant contraction of the adductor pollicis
muscle was recorded using a force displacement
transducer and neuromuscular function analyser
(Myograph 2000, Biometer International, Odense,
Denmark). Preload tension on the thumb was
maintained at 300 £ throughout the investigation.
The first twitch (Tl) of the TOF was considered the
twitch height.
After end-tidal anaesthetic concentration had been
stable for 30 min, 96 patients were allocated
randomly to receive atracurium, mivacurium or a
combination of atracurium and mivacurium. The
following doses of drugs were administered to
subgroups of eight patients: atracurium 20, 60, 80,
100, 120 ug kg"1; mivacurium 10, 20, 40, 60 ug kg"1.
Studies of the single drug groups were concluded
first, so that doses of combinations could be planned.
From the dose-response curves of the neuromuscu-
lar agents administered alone, the respective ED,o
values (effective dose resulting in 50 % reduction in
Tl) were determined. Subsequently, a dose-
response curve was obtained by administration of
the two drugs in a constant dose ratio based
on the ED^ values of the single agent. The
following dosage combinations were administered:
1 EDJO atracurium+1 EDW mivacurium; 0.5
atracurium + 0.5 ED^ mivacurium; and 0.25
atracurium + 0.25 ED^ mivacurium). From the
dose-response curve of the combined drugs, the
ED M value of the total dose of the mixture was calcu-
lated and, based on the known dose ratio, the single
doses of the agents in the combination were obtained
for plotting on the isobologram. All drugs were ad-
485
ministered randomly to subgroups of eight patients
and were injected over 5 s into a rapidly flowing i.v.
infusion. In the combination group, drugs were
injected simultaneously into two separate i.v.
catheters inserted in one arm. The neuromuscular
response was recorded as the maximum depression
of T l , expressed as a percentage of the control value.
When the maximum effect of the selected dose was
reached (that is, when no further decrease in evoked
response to three consecutive stimuli occurred), the
study was terminated and anaesthesia continued as
appropriate for surgery.
Data processing and isobolographic analysis
The percentage values for Tl depression in each
group were transformed to probits and plotted
against the logarithm of the dose using PCNONLIN
version 4.2A (ClinTrials, Inc., Lexington, KY,
USA) [12]. Regression lines were compared using
analysis of covariance. First, we tested the lines to
determine if they deviated from parallelism. If they
did not, an F test was applied to determine if the
elevations were different. If so, a t test was applied to
determine which line differed in elevation [13], using
BMDP statistical package, release 7.0 (University of
California Press, Berkeley, CA, USA, 1993). The
ED-jo and EDW values (doses causing 50 % and 95 %
depression of twitch tension, respectively) were
calculated from the log-probit regression lines for
each group. Using analysis of variance, we compared
age and body weight, between different groups.
Unless otherwise specified, the results are expressed
as mean (95 % confidence intervals (CI)) and were
considered significant when P < 0.01.
Isobolographic [10,11] and algebraic (fractional)
[14] analyses were used (EDK level) to define the
type of interaction between atracurium and miva-
curium. Isobolographic analysis for drug-drug inter-
actions was conducted according to the procedure of
Tallarida, Porreca and Cowan [10]. This analysis has
the advantage of being independent of the slopes of
the dose-response curves that is, parallelism does
not have to be established. The isobolograms were
constructed by plotting single drug ED^, points on
the dose co-ordinates of the isobologram and a
combined ED^ point in the dose field. A straight line
joining the single drug ED^ points is termed an
isobole (isos, equal, bole, effect) or the "additive
line". The entire graph is termed an isobologram. If
the EDJO of a combination falls on the theoretical
additive line, the effect of the drug mixture is
additive (zero interactive). Points to the left of the
theoretical additive line would be consistent with a
synergistic interaction, whereas points to the right of
the line would indicate an antagonistic interaction.
CI for each point were calculated from the variances
of each component alone and were evaluated for
statistical significance using the Student's t test.
Algebraic analysis was used to describe the
magnitude of the interaction. It was based on the
expression of the component doses of the two agents
of the combination as fractions of the doses that
produce the same effect when given separately. The
486
sum of the fractional doses, as expressed by the
following equation, indicates the type of interaction:
where (EDW), and (ED,o)m = EDW values of atra-
curium and mivacurium, respectively, given alone,
and da and dm doses of atracurium and mivacurium,
respectively which, when combined, are equipotent
with (EDSQ), or (ED,,,),,,. Values near 1 indicate
additive interaction, values greater than 1 imply an
antagonistic interaction and values less than 1
indicate a synergistic interaction.
CLINICAL STUDIES
Selection of patients, premedication and intra-
operative monitoring were similar to that described
in the isobolographic study. Anaesthesia was induced
with fentanyl 2 ug kg"1 and thiopentone 5 mg kg"1,
and maintained with 60% nitrous oxide and iso-
flurane in oxygen. The trachea was sprayed with 4 %
lignocaine 4 ml and intubated without the use of
neuromuscular blockers. After intubation, the end-
tidal concentration of isoflurane was adjusted to
1.2% (1 MAC excluding nitrous oxide). After the
end-tidal anaesthetic concentration had been stable
for 30 min and after establishment of a stable
neuromuscular response, patients were allocated
randomly to two groups to receive either atracurium
0.5 mg kg"1 (n = 13) or mivacurium 0.15 mg kg"1
(« = 13) i.v. Additional maintenance doses of miva-
curium 0.1 mg kg"1 were administered to patients in
both groups, whenever the first twitch recovered to
10 % of control. The onset time (time from the end
of injection to complete suppression of response to
TOF stimulation) and times from end of injection to
5% and 10% recovery of T l (after the initial and
maintenance doses), were noted in both groups. At
the end of surgery, residual neuromuscular block
was antagonized with neostigmine 50 ug kg"1 when
T l had recovered to 10 % of its initial control height.
Atropine lSugkg" 1 was administered with neo-
stigmine to all patients. Statistical analysis was
carried out using Student's t test. Differences
were considered significant when P ^ 0.01.
Results
INTERACTION STUDIES
The calculated doses for the ED W values of the first
twitch height were 50.5 (95% CI 48.9-52.1) and
20.8 (20.3-21.3) ug kg"1 for the atracurium and
mivacurium groups, respectively. Corresponding
ED,, values were 103.9 (100.3-107.7) and 37.4
(35.8-39) ug kg"1, respectively. The dose-response
curve of each drug was plotted against the logarithm
of the dose in EDW units (fig. 1). The regression lines
of mivacurium, atracurium and the atracurium—
mivacurium combination did not deviate from
parallelism.
Isobolographic analysis demonstrated an additive
interaction with respect to the neuromuscular
blocking activity of the atracurium—mivacurium
combination. Figure 2 represents the EDW isobolo-
gram for the atracurium-mivacurium interaction.
British Journal of Anaesthesia
The experimentally determined ED^ value for the
combination was 23.9 (22.4-25.3) ug kg"1 for atra-
curium and 9.9 (9.3-10.5) |ig kg"1 for mivacurium.
The theoretical additive ED^ value was calculated as
25.3 (23.9-26.6) ug kg"1 for atracurium and 10.4
(9.9-10.9) ug kg"1 for mivacurium. The confidence
intervals of these points overlap and the result of
Student's t test for potency ratio was statistically
insignificant, indicating zero interaction (additivism).
The fractional (algebraic) analysis of this interaction
also demonstrated additivism (table 1).
CLINICAL STUDIES
Complete neuromuscular block was achieved in all
patients who received either atracurium 0.5 mg kg"1
or mivacurium 0.15 mg kg"1. There was no difference
in the onset time of the initial dose of the neuro-
muscular blocking agents in the two groups: 59.4
(42.2-76.6) and 81.8 (63.8-99.9) s for the atracurium
and mivacurium groups, respectively (table 2).
Times to 5% and 10% recovery of Tl after
atracurium 0.5 mg kg"1 were significantly (P <
0.0005) longer than those observed after mivacurium
0.15 mg kg"1 (table 2).
There was a marked difference in the action of the
first maintenance dose of mivacurium according to
99-i
98-
95-
90-
|80
o 70
60-
50-
•0-
8 30
10'
5-
2-
1 -
0.3 0.5 1.0 2.0 3.0
E D M equivalents
Figure 1 Log dosc-probit plot for the first twitch depression
for atracurium (#), mivacurium ( • ) and their combination
(A)- Individual points represent mean depression of first twitch
height (Tl) (% control) with each dose and bars represent 95 %
confidence intervals. Drug doses are represented as EDK
equivalents.
A tracurium-mivacurium interactions 487

25 T (table 2). The duration of the first and second

maintenance doses of mivacurium to 5% and 10%
recovery are presented in table 2.

Discussion
20 -
In the current study, isobolographic analysis demon-
strated that the combination of atracurium and
mivacurium exerted no greater effect than that seen
with either agent administered alone. The fractional
(algebraic) analysis of this interaction also demon-
strated zero (additive) interaction (table 1) that is the
3. effect of the combination was precisely that expected
from the dose-response relations of the individual
agents. The results of this study are consistent with

A
3
u
.5 10
ED^forA+M \ produce an additive response [6,7]. Atracurium
5-
0 10 20 30
Atracurium (/jg kg"1)
Figure 2 Tl EDJQ isobologram for the interaction of
atracurium (A) and mivacurium (M). The dashed line
connecting the single drug ED M points is the theoretical
additive line and the point on this line is the theoretical
additive EDJO point (95 % confidence intervals). The
experimentally determined EDW dose (95 % confidence
intervals) of the atracurium-mivacurium combination could not
be distinguished from simple additive action.
which neuromuscular blocking agent was ad-
ministered first. The average duration of action of
the first supplementary dose of mivacurium was 1.8
times longer after atracurium than after mivacurium
the observation that combinations of structurally
similar neuromuscular blocking drugs (pancuronium
and vecuronium or pipecuronium and vecuronium)
and mivacurium have the same chemical structure;
both are bis-benzylisoquinolinium non-depolarizing
neuromuscular block agents.
\ ED^forA Our results indicated that the use of mivacurium
after atracurium led to increased duration of the first
supplementary dose of mivacurium (table 2). With
subsequent increments, there was no significant
difference in the duration of action of the second
40 50 maintenance dose of mivacurium whether it was
administered after atracurium or mivacurium (table
2). In this respect, our results are in agreement with
those of other investigators [15-18]. The use of
incremental doses of atracurium or vecuronium after
recovery of tubocurarine block to 10% of control of
the first twitch was studied by Middleton and
colleagues [15]. The result was that the intensity and
duration of the neuromuscular block after the first
increment was always greater than that of subsequent
increments. They also noted [15] that the duration
and intensity of the block was reduced progressively
with subsequent increments until the response to
further increments was unchanged.

Table 1 Atracurium-mivacurium interaction. Equipotent doses (EDM) of the atracurium-mivacurium

combination
Atracurium component Mivacurium component

Fraction of Dose Fraction of Dose Sum of

Group ED W (ug kg"1) ED M (Ug kg"1) fractions
Atracurium 1 50.5 1
Mivacurium 1 20.8 1
Combination 0.47 23.8 0.47 9.8 0.94

Table 2 Mean (95 % confidence intervals) onset and duration of action of atracurium 0.5 mg kg"1 and mivacurium 0.15 mg kg"1, and
duration of maintenance doses of mivacurium 0.1 mg kg"1. *** Significantly different (P < 0.001) from the corresponding value of
mivacurium
Maintenance doses
Fun Second
Onset time Time to 5% Tl Tune to 10% Tl Tune to 5% Tl Tune to 10% Tl Tune to 5% Tl Time to 10% Tl
Group recovery (min) recovery (min) recovery (min) recovery (min) rccorcfy (min) recovery (min)

Atr»curhm) 59.4 (42.2-76.6) 37.6 (32.9-42.4)*** 40.5 (35.6-453)*** 23.5 (20.1-26.9)*** 25.2 (21-8-28.5)*•* 16.7 (11.5-21.9) 183 (12.6-24)
0.5 mg kg
Mivscurium 81.8 (63.8-99.9) 15.9 (13.7-18.3) 17.1 (14.9-19.4) 12.9 (10.5-15.1) 14.2 (11.9-16 6) 13.8 (9.95-17.8) 14.6 (10.6-18.6)
488
Similarly, it has been shown that the effect of a
maintenance dose of vecuronium given during
recovery from an initial dose of either pipecuronium
[16] or pancuronium [17,18] is dependent on the
neuromuscular blocker which was used initially.
Smith and White [16] noted that the duration of the
vecuronium maintenance dose was greater (P =
0.02) after pipecuronium (40 (SD 12) min) than after
vecuronium (29 (9) min) and was similar to that of
pipecuronium after pipecuronium (49 (15) min).
After pipecuronium, they detected clinically signifi-
cant prolongation after four maintenance doses of
vecuronium [16]. After pancuronium, Kay and
colleagues [17] noted that the first two maintenance
doses of vecuronium were prolonged; however, this
effect was negligible by the third dose. Further,
Rashkovsky, Agoston and Ket [18] have shown that
if, at the time of 25% recovery from complete
pancuronium neuromuscular block, 95 % paralysis is
re-established with incremental doses of vecuronium,
the required vecuronium dose and the duration of
paralysis is dependent on the neuromuscular blocker
that was used initially.
The apparent prolongation of action of the first
maintenance dose of mivacurium administered after
atracurium in this study and those reported with
vecuronium in the aforementioned studies [15—18]
was not related to synergism because combinations
of atracurium and mivacurium (in this study) and
combinations of vecuronium and pancuronium or
pipecuronium are simply additive [6,7]. However,
this prolongation in the duration of action could be
attributed to the relative concentrations of these
drugs at the receptor site. As the majority of receptors
remain occupied by the drug administered initially,
the clinical profile depends on the kinetics of the
drug administered first rather than on that of the
second (maintenance) drug. However, with further
incremental doses of the second drug, a progressively
larger proportion of the receptors is occupied by the
second drug and the kinetics (the clinical profile) of
that drug become evident.
In this study, the calculated first twitch ED*, and
ED,, values for atracurium during thiopentone—
fentanyl-nitrous oxide-isoflurane anaesthesia were
50.5 (48.9-52.1) and 103.9 (100.3-107.7) ug kg"1,
respectively. Similar values (70 and 130ugkg~',
respectively) were reported by Sokoll and colleagues
[19] and by Rupp, Fahey and Miller [20] (68 and
122 ug kg"1, respectively) during isoflurane (0.60-
0.75% end-tidal) and 60% nitrous oxide anaes-
thesia. The ED.* and ED,, values of 20.8 (20.3-21.3)
and 37.4 (35.8-39) ug kg"1, respectively, for miva-
curium calculated in the present study are in close
agreement (29 and 45 ug kg"1, respectively) with
those reported by Weber and colleagues [21] who
used train-of-four mode of stimulation and mech-
anomyography during isoflurane (0.50-0.75%
end-tidal) and nitrous oxide anaesthesia.
Volatile anaesthetics potentiate the effects of
neuromuscular blocking agents and this effect has
been attributed to increased muscle blood flow,
presynaptic inhibition of acetylcholine mobilization
and release, postsynaptic receptor desensitization
and an action on the muscle at some point distal to
British Journal of Anaesthesia
the cholinergic receptor [22,23]. Therefore, it is
expected that isoflurane augments both atracurium-
and mivacurium-induced neuromuscular block. This
can be illustrated by comparing our results with
those determined in previous studies during an-
aesthesia without a volatile anaesthetic. Sokoll and
co-workers reported that the EDM and ED^ of
atracurium during nitrous oxide and fentanyl an-
aesthesia were 120 and 270 ug kg"1, respectively [19].
These are approximately 2.5 times the values
we determined for EDW and EDW (50.5 and
103.9 ug kg"1, respectively). Similarly, Weber and
colleagues [21] and Naguib [24] reported that the
ED^ and EDW values of mivacurium were 41 and
58ngkg~', and 37.1 and 68.5 ug kg"1, respectively,
during nitrous oxide and fentanyl anaesthesia.
These are approximately 1.5-1.9 times the values
we determined for EDTO and ED^ (20.8 and
37.4 ug kg"1, respectively). The aforementioned dif-
ferences are consistent with those found by other
investigators for comparisons of isoflurane with
balanced anaesthesia [19,21].
References
1. Savarese JJ, Ali HH, Basra SJ, Embree PB, Scott RPF,
Sunder N, Weakly JN, Wastila WB, El-Sayad HA. The
clinical neuromuscular pharmacology of mivacurium chloride
(BW B1090U). A short-acting nondepolarizing ester neuro-
muscular blocking drug. Anesthesiology 1988; 68: 723-732.
2. Ah' HH, Savarese JJ, Embree PB, Basta SJ, Stout RG,
Bottros LH, Weakly JN. Clinical pharmacology of miva-
curium chloride (BW B1090U) infusion: comparison with
vecuronium and atracurium. British Journal of Anaesthesia
1988; 61: 541-546.
3. Katz RL. Modification of the action of pancuronium by
succinylcholine and halothane. Anesthesiology 1971; 35:
602-606.
4. Young RB. Suxamethonium for peritoneal closure. An-
aesthesia 1979; 34: 716.
5. Waud BE, Waud DR. Quantitative examination of the
interaction of competitive neuromuscular blocking agents on
the indirectly elicited twitch. Anesthesiology 1984; 61:
420-427.
6. Ferres CJ, Mirakhur RK, Pandit SK, Clarke RSJ, Gibson
FM. Dose—response studies with pancuronium, vecuronium
and their combination. British Journal of Climcal Phar-
macology 1984; 18: 947-950.
7. Naguib M, Abdulatif M. Isobolographic and dose-response
analysis of the interaction between pipecuronium and
vecuronium in surgical patients. British Journal of Anaesthesia
1993; 71: 556-560.
8. Lebowitz PW, Ramsey FM, Savarese JJ, Ah' HH. Potentia-
tion of neuromuscular blockade in man produced by combina-
tions of pancuronium and metocurine or pancuronium and d-
rubocurarine. Anesthesia and Analgesia 1980; 59: 604-609.
9. Mirakhur RK, Gibson FM, Fcrrcs CJ. Vecuronium and d-
tubocurarine combination: potentiation of effect. Anesthesia
and Analgesia 1985; 64: 711-714.
10. Tallarida RJ, Porreca, F, Cowan A. Statistical analysis of
drug-drug and site-site interactions with isobolograms. Life
Sciences 1989; 45: 947-961.
11. Berenbaum MC. What is synergy? Pharmacological Reviews
1989; 41: 93-141.
12. Statistical Consultants, Inc. PCNONLIN and NONLIN84.
Software for the statistical analysis on nonlinear models.
American Statistician 1986; 40: 52.
13. Finney DJ. Statistical Methods in Biological Assay, 2nd Edn.
London: Griffin, 1971; 21-57.
14. Berenbaum MC. Synergy, addin'vism and antagonism in
immunosuppression. Journal of Clinical and Experimental
Immunology 1977; 28: 1-18.
15 Middleton CM, Pollard BJ, Healy TEJ, Kay B. Use of
Atracurium-mivacurium interactions
atracurium or vecuronium to prolong the action of tubo-
curarine. British Journal of Anaesthesia 1989; 62: 659-663.
16. Smith I, White PF. Pipecuronium-induced prolongation of
vecuronium neuromuscular block. British Journal of An-
aesthesia 1993; 70: 446-148.
17. Kay B, Chestnut RJ, Sum Ping JST, Healy TEJ. Economy in
the use of muscle relaxants. Vecuronium afteT pancuronium.
Anaesthesia 1987; 42: 277-280.
18. Rashkovsky OM, Agoston S, Ket JM. Interaction between
pancuronium bromide and vecuronium bromide. British
Journal of Anaesthesia 1985; 57: 1063-1066.
19. SokoU MD, Gergis SD, Mehta M, Ah NM, Lineberry C.
Safety and efficacy of atracurium (BW33A) in surgical
patients receiving balanced or isoflurane anesthesia. Anes-
thesiology 1983; 58: 450-455.
20. Rupp SM, Fahey MR, Miller RD. Neuromuscular and
cardiovascular effects of atracurium during nitrous oxide—
489
fentanyl and nitrous oxide—isoflurane anaesthesia. British
Journal of Anaesthesia 1983; 55 (Suppl. I): 67S-70S.
21. Weber S, Brandom BW, Powers DM, Sarner JB, Woelfel
SK, Cook RD, Foster VJ, McNulty BF, Weakly JN.
Mivacurium chloride (BW B1090U)-induced neuromuscular
blockade during nitrous oxide—isoflurane and nitrous oxide-
narcotic anesthesia in adult surgical patients. Anesthesia and
Analgesia 1988; 67: 495-^99.
22. Waud BE, Waud DR. Comparison of the effects of general
anesthetics on the end-plate of skeletal muscle. Anesthesiology
1975; 43: 540-547.
23. Waud BE, Waud DR. The effects of diethyl ether, enflurane,
and isoflurane at the neuromuscular junction. Anesthesiology
1975; 42: 275-280.
24. Naguib M. Neuromuscular effects of rocuronium bromide
and mivacurium chloride administered alone and in com-
bination. Anesthesiology 1994; 81: in press.