Full text of DOI:10.1592/phco.22.11.823.33625

_
B RIEF R EP RTS
Tachyphylaxis Associated with Continuous Cisatracurium
versus Pancuronium Therapy
Salmaan Kanji, Pharm.D., Jeffrey F. Barletta, Pharm.D., James J. Janisse, Ph.D.,
James A. Kruse, M.D., FCCM, and John W. Devlin, Pharm.D.
Study Objectives. To compare dosing requirements over time among patients
receiving continuous cisatracurium versus pancuronium therapy, and to
identify factors that may account for changes in pancuronium versus
cisatracurium infusion requirements over time.
Design. Retrospective, comparative cohort analysis.
Setting. A tertiary level 1 trauma center.
Patients. Forty-five consecutive adult patients who were admitted to
intensive care units at our institution from January 1998–August 2000 and
received continuous cisatracurium or pancuronium therapy for at least 48
hours.
Measurements and Main Results. Dosing requirements of patients treated
with pancuronium or cisatracurium were recorded over time throughout
the treatment period. Factors that could affect dosing requirements of a
neuromuscular blocking agent (NMBA) were stratified as time invariant
(admitting service, acute physiology and chronic health evaluation II score,
duration of mechanical ventilation, pressure control ventilation, baseline
hepatic or renal insufficiency, thermal injury, train-of-four assessment, and
concurrent drug administration or disorders affecting neuromuscular
transmission) or time variant (concurrent sedation and narcotic analgesia
therapy; serum magnesium, potassium, and creatinine concentrations;
arterial pH level; temperature; peak airway pressure; and partial pressure of
oxygen:fraction of inspired oxygen ratio). Hierarchical linear modeling was
used to compare the dosing requirements and to identify confounders
affecting the relationship. The infusion rate escalation for the
cisatracurium group was greater (0.39 µg/kg/min; 95% confidence interval
[CI] 0.22–0.56; 23 patients) than for the pancuronium group (-0.06
µg/kg/min; 95% CI -0.24–0.12; 22 patients; p<0.001) and was associated
with an average daily cost/patient significantly higher (p<0.001) with
cisatracurium ($258 ± $114) than pancuronium ($11 ± $5). Confounder
analysis revealed that only the admitting service and the number of times
the NMBA infusion was suspended because no twitch was detected differed
between groups. Neither of these confounders significantly affected the
temporal relationship between cisatracurium and pancuronium infusion
rates.
Conclusion. Dosing requirements increase over time at a significantly greater
rate for cisatracurium than pancuronium infusions. Tachyphylaxis with
cisatracurium is associated with substantial drug-related costs and is not
accounted for by various disease-, patient-, and therapy-related factors.
Further investigation is required to elucidate the mechanisms and risk
factors underlying this phenomenon.
(Pharmacotherapy 2002;22(7):823–830)

824
PHARMACOTHERAPY Volume 22, Number 7, 2002
Continuous therapy with nondepolarizing
between patients in the ICU receiving continuous
pancuronium versus cisatracurium therapy in
order to identify potential tachyphylaxis with
cisatracurium. The secondary objective was to
identify factors that might account for changes in
pancuronium or cisatracurium infusion
requirements over time.
neuromuscular blocking agents (NMBAs) is
common in intensive care units (ICUs) to
facilitate mechanical ventilation and promote gas
exchange in patients not responding to sedation
therapy alone.1 Pancuronium (Pavulon;
Organon, West Orange, NJ) is a long-acting
aminosteroid NMBA that has been recommended
as first-line therapy for most patients who require
continuous NMBA therapy.2 Cisatracurium
(Nimbex; Glaxo Wellcome, Research Triangle
Park, NC) is a benzylisoquinolinium NMBA that
often is substituted for pancuronium when safety
is a concern with pancuronium administration.
Unlike pancuronium, cisatracurium has a
clearance mechanism independent of organ
function, has not been associated with vagolytic
effects, and is less likely to cause prolonged
muscle weakness.3–5 Although cisatracurium has
a potential safety advantage over pancuronium,
its cost is almost 10 times greater.6 For this
reason, continuous NMBA treatment guidelines
at our tertiary level 1 trauma center, which are
consistent with recently published clinical
practice guidelines,2 restrict administration of
cisatracurium to patients with renal insufficiency,
hemodynamically significant increases in heart
rate related to pancuronium therapy, concurrent
corticosteroid administration, or a history of
asthma or bronchospasm.
Methods
This retrospective cohort study was approved
by the Wayne State University Human
Investigation Committee. Hospital records were
reviewed for 45 consecutive adult patients who
were admitted to intensive care units (both
medical and surgical) at Detroit Receiving
Hospital from January 1998–August 2000 and
received continuous cisatracurium or
pancuronium therapy for at least 48 hours. The
hospital pharmacy computer system was used to
identify patients. The attending physician
decided whether continuous NMBA therapy was
to be administered. Continuous NMBA therapy
at our hospital is administered to facilitate
mechanical ventilation in patients with
excessively high airway pressure, ventilator
asynchrony, or poor oxygenation despite
ventilator manipulation and therapeutic sedation.
Hospital policy mandates that all patients
receiving continuous NMBA therapy must receive
concomitant sedation and narcotic analgesia
therapy and that train-of-four monitoring with a
peripheral nerve stimulator be performed at least
every 6 hours. The NMBA infusion rate must be
decreased if no twitch is detected. In our
analysis, therefore, we assumed that sedation and
narcotic analgesia therapy was performed most
effectively by the nurse, using physiologic
indicators such as heart rate, blood pressure, and
respiratory rate; that NMBA infusion rates were
titrated quickly to the lowest effective dosage;
and that NMBA dosing was determined most
effectively using both train-of-four monitoring
and clinical evaluation.
Anecdotally, several patients in the ICU who
received continuous cisatracurium therapy at our
institution required progressively higher infusion
rates over their course of therapy despite an
adequate clinical response at the start of therapy.
This dosage escalation has not been observed
with pancuronium. Rapid, diminished, or failed
pharmacologic response to a drug, despite
continued administration, is known as
tachyphylaxis or resistance.7 Tachyphylaxis has
been described with many NMBAs, including
pancuronium,8, 9 vecuronium,10–16 atracurium,13,
and rocuronium.25 Although tachyphylaxis
17–24
has been observed with atracurium, it has not
been reported in patients in the ICU with
atracurium’s R-cis, R′-cis isomer, cisatracurium.
The primary objective of this cohort analysis
was to compare infusion requirements over time
Indication for continuous NMBA therapy was
not an inclusion criterion. Patients admitted
with cranial or spinal trauma were excluded.
Data were collected for each patient until
continuous NMBA therapy was discontinued
because the patient improved or a different
NMBA regimen was started, or the patient died.
In each instance, we assumed that the clinician
rapidly increased the dosage of either
cisatracurium or pancuronium until the desired
clinical end point of NMBA therapy was reached.
From the College of Pharmacy (Dr. Kanji) and the Center
for Health Effectiveness Research (Dr. Janisse), Wayne State
University; and the Departments of Pharmacy Services (Drs.
Barletta and Devlin) and Medicine (Dr. Kruse), Detroit
Receiving Hospital, Detroit, Michigan.
Address reprint requests to John W. Devlin, Pharm.D.,
BCPS, Department of Pharmacy Services, Detroit Receiving
Hospital, 4201 St. Antoine, Detroit, MI 48201.

TACHYPHYLAXIS ASSOCIATED WITH CONTINUOUS CISATRACURIUM THERAPY Kanji et al 825
Patient records indicated that the average
infusion rate and total dose of cisatracurium or
pancuronium administered were determined at 6-
hour intervals. Average infusion rates were
calculated by dividing the cumulative weight-
adjusted dose by the 6-hour interval.
Cisatracurium and pancuronium costs were
calculated using institutional contract pricing.6
Costs related to NMBA preparation and
administration were not considered.
A literature review using MEDLINE was
conducted from January 1966–August 2000 to
identify both patient and nonpatient variables
that might affect cisatracurium or pancuronium
infusion requirements. Possible confounders
were categorized as either time invariant or time
variant. Time-invariant confounders were age,
weight, gender, admitting service, acute
physiology and chronic health evaluation
(APACHE) II scores26 on admission to ICU and at
the start of continuous NMBA therapy, length of
stay, survival to hospital discharge, duration of
mechanical ventilation, prolonged (> 24 hrs)
pressure control ventilation, length of time to
wean from mechanical ventilation, baseline
hepatic dysfunction (defined as serum total
bilirubin concentration > 5 mg/dl or non-
pharmacologic prolongation of international
normalized ratio > 2), baseline renal insufficiency
(defined as creatinine clearance < 30 ml/min
estimated by a published method27), thermal
injury, number of train-of-four assessments that
resulted in no detectable twitch, and concomitant
medical disorders or drug administration that
could affect NMBA infusion requirements.
Confounders were considered time variant if
sequential measurements were made routinely
over the course of NMBA therapy. Respiratory
rate, heart rate, and concomitant morphine and
lorazepam administration were recorded at 6-
hour intervals. Maximum body temperature;
mode and rate of mechanical ventilation; peak
airway pressure and peak end-expiratory
pressure; serum magnesium, potassium, and
creatinine concentrations; arterial pH; and
arterial oxygen tension (torr):fraction of inspired
oxygen ratio (PaO2:FiO2) were recorded for every
24-hour period of cisatracurium or pancuronium
administration.
Initially, univariate analyses of the confounders
(both time invariant and time variant) were
conducted to identify differences between the
cisatracurium- and pancuronium-treated patient
groups. Continuous time-invariant confounders
were compared using Student’s t test for
independent samples. The Mann-Whitney U test
was used for continuous, nonparametric data.
Dichotomous, time-invariant variables were
2
compared using the ␹ statistic with the Yates
continuity correction. Time-variant confounders
were compared between groups using individual
hierarchical linear models for each possible
confounder. A p value less than 0.05 (2-tailed
test) indicated statistical significance. Finally,
Pearson correlation coefficients were calculated
for each patient between time-variant
confounders and NMBA dosage so that the
pattern of intrapatient correlations between each
time-variant covariate and NMBA dosage could
be examined.
Since duration of therapy differed for each
patient, standard approaches using repeated-
measures analysis of variance could not be
applied appropriately to compare cisatracurium
and pancuronium infusion rates over time.
Instead, we used hierarchical linear modeling
(HLM version 4.04, 1998, Scientific Software
International Inc., Chicago, IL), which can
account for various treatment durations.28
Possible covariates for this analysis were any of
the time-invariant and time-variant confounders
that were related to drug group membership at a
p value of 0.10 or less from the univariate
analyses. Once inserted in the hierarchical linear
model, confounders were retained if their
probability level was less than 0.05.
Results
A total of 45 patients met the inclusion criteria
for analysis; 23 received cisatracurium and 22
received pancuronium. The only time-invariant
confounders that differed between the
cisatracurium and pancuronium groups were the
admitting service and the number of times the
NMBA infusion was suspended because no
twitch was detected with the train-of-four
assessment (Table 1). Patients receiving
cisatracurium were more likely to be admitted to
a medical service, whereas patients receiving
pancuronium were more likely to be admitted to
a surgical service (p=0.021). More pancuronium-
treated patients were likely to have NMBA
infusions suspended because no twitch was
detected (p=0.033). Univariate analysis of time-
variant covariates revealed no significant
differences between groups (Table 2). The
pattern of intrapatient correlations between each
time-variant covariate and NMBA dosage
revealed a substantial level of heterogeneity. For

826
PHARMACOTHERAPY Volume 22, Number 7, 2002
Table 1. Univariate Analysis of Potential Time-Invariant Confounders
Cisatracurium
Pancuronium
(n=22)
42.5 (21–62)
84 ± 19
18/4
(n=23)
p Value
Age, yrsa
42 (25–77)
81 ± 23
15/8
0.400c
0.645d
0.208e
0.021e
0.312d
0.558d
0.413c
0.510c
0.181e
0.547c
Weight, kgb
M/F
Admitting service, no. medical/surgical
APACHE II score at admissionb
APACHE II score at start of NMBA therapyb
Length of hospital stay, daysa
13/10
18 ± 7
20 ± 7
25 (7–65)
20 (7–60)
9 (39)
5/17
15 ± 8
19 ± 6
19 (4–93)
16 (2–86)
13 (59)
14.5 (2–86)
Length of ICU stay, daysa
Survival, no. (%) of pts
Duration of mechanical ventilation, daysa
Required pressure control ventilation for > 24 hrs,
(%) of pts
20 (5–56)
13 (57)
1 (4)
5 (22)
75 ± 41
5 (22)
10 (44)
0 (0)
2 (9)
94 ± 33
4 (18)
0.555d
0.511f
0.414f
0.086d
0.530f
Hepatic dysfunction, no. (%) of pts
Renal insufficiency, no. (%) of pts
Creatinine clearance, ml/minuteb
Thermal injury, no. (%) of pts
NMBA infusion suspended because no twitch
detected with train-of-four (no. of times)a
Concurrent disorder that interferes with
neuromuscular transmission,g no. (%) of pts
Concurrent drug administration that interferes
with neuromuscular transmission,h no. (%) of pts
0 (0–2)
11 (48)
17 (74)
1 (0–4)
8 (36)
0.033c
0.436e
0.457e
14 (64)
NMBA = neuromuscular blocking agent, APACHE = acute physiology and chronic health evaluation; ICU = intensive care unit.
aData are median (range).
bData are mean ± SD.
cMann-Whitney U test.
dStudent’s t test.
e
␹2 statistic with Yates correction.
fFisher’s exact test.
gDisorders were multiple sclerosis, cerebral palsy, amyotrophic lateral sclerosis, Guillain-Barré syndrome, muscular dystrophy,
myasthenia gravis, Eaton-Lambert syndrome, peripheral neuropathy, spinal cord injury, sepsis, or cancer.
hDrugs were anesthetic agents, aminoglycosides, clindamycin, vancomycin, antidysrhythmics, steroids, lithium, xanthine derivatives,
dantrolene, cyclosporine, anticonvulsants, or anticholinesterase agents.
Table 2. Analysis of Potential Time-Variant Confounders
Cisatracurium Group
(n=23)
Pancuronium Group
(n=22)
Morphine (mg/hr)a
Lorazepam (mg/hr)a
0.67 (0–8.0)
2.0 (0–6.0)
1.1 (0–8.4)
2.0 (0.8–9.0)
Magnesium (mEq/L)b
Potassium (mEq/L)b
Creatinine (mg/dl)a
2.1 ± 0.4
4.0 ± 0.5
1.0 (0.4–11.3)
2.2 ± 0.5
4.2 ± 0.4
1.0 (0.5–8.2)
Arterial pH levelb
7.38 ± 0.06
38.1 ± 1.0
35 ± 7
7.39 ± 0.05
38.2 ± 0.9
36 ± 9
Body temperature (°C)b
Peak airway pressure (mm Hg)b
PaO2:FiO2 ratiob
134 ± 40
181 ± 69
PaO2:FiO2 = arterial oxygen tension (torr):fraction of inspired oxygen.
aData are median (range).
bData are mean ± SD.
Individual hierarchical linear models demonstrated that probability for each comparison was
not significant.
many of the confounders, the correlation
coefficients between patients ranged from almost
1.0 to -1.0.
Results of the hierarchical linear modeling
analysis indicated that increases in infusion
rates/day of therapy were significantly greater in
the cisatracurium group (0.39 µg/kg/min, 95%
confidence interval [CI] 0.22–0.56) than in the
pancuronium group (-0.06 µg/kg/min, 95% CI
-0.24–0.12, p<0.001). As expected, the average

TACHYPHYLAXIS ASSOCIATED WITH CONTINUOUS CISATRACURIUM THERAPY Kanji et al 827
Table 3. Comparison of Cisatracurium and Pancuronium Infusion Rates
Cisatracurium Group
(n=23)
0.39 (0.22–0.56)
2.17 (1.77–2.47)
174 (54–360)
412 ± 183
Pancuronium Group
(n=22)
-0.06 (-0.24–0.12)
0.88 (0.46–1.30)
143 (48–534)
88 ± 39
p Value
Change in daily infusion rate (µg/kg/min)a
Starting infusion rate (µg/kg/min)a
Duration of treatment (hrs)c
Daily dose (mg)e
<0.001b
<0.001b
0.052d
<0.001f
aData are mean (95% confidence interval).
bHierarchical linear modeling analysis.
cData are median (range).
dMann-Whitney U test.
eData are mean ± SD.
fStudent’s t test.
starting infusion rate of cisatracurium (2.17
µg/kg/min, 95% CI 1.77–2.57) was significantly
greater than that of pancuronium (0.88
µg/kg/min, 95% CI 0.46–1.30, p<0.001) (Figure
1, Table 3). The average daily cost was
significantly higher (p<0.001) for patients treated
with cisatracurium ($258 ± $114) than
pancuronium ($11 ± $5).
Based on the univariate analysis, only two
possible confounders (admitting service and
number of times the NMBA infusion was not
administered because no twitch was detected)
were considered in the hierarchical linear
modeling analysis that compared infusion rates
of cisatracurium and pancuronium over time.
The hierarchical linear modeling analysis found
that neither of these covariates met the retention
criterion (p<0.05).
Discussion
Tachyphylaxis associated with continuous
neuromuscular blockade has been reported with
other benzylisoquinolonium-structured NMBAs,
such as atracurium, but not with cisatracurium.13,
17–24
This is the first large case series using a
comparator group and an extensive confounder
analysis to investigate factors that could account
for this phenomenon. Our findings demonstrate
that increasingly higher dosages are required over
time by patients treated with continuous
cisatracurium versus pancuronium. The
confounder analysis suggests that this increase
was not attributable to clinical deterioration or
other factors that affect NMBA dosing requirements.
Our primary analysis demonstrated that the
average daily infusion rate of cisatracurium
increased by 0.39 µg/kg/minute. However,
variability was significant with respect to the
magnitude of the infusion rate increase among
these patients. To determine if the observed
increased infusion rate with cisatracurium
resulted from a large increase in cisatracurium
dosing in a few patients, the magnitude of dosage
increase for each patient receiving cisatracurium
was examined individually. A daily infusion-rate
increase of more than 0.75 µg/kg/minute was
deemed consistent with clinically significant
tachyphylaxis. Six of the 23 cisatracurium-
treated patients met this criterion. The median
increase in the daily infusion rate for these six
patients was 0.88 µg/kg/minute (range 0.75–1.76
µg/kg/min). This response was likely responsible
for the significant dosing increase observed with
cisatracurium in the hierarchical linear modeling
analysis. However, on reanalysis that excluded
the data for these six patients, the dosing rate for
20
Pancuronium
Cisatracurium
15
10
5
0
0
10
20
30
40
50
60
Time in 4-Hour Intervals
Figure 1. Change in pancuronium (solid line) and
cisatracurium (dotted line) infusion rates over time
generated by hierarchical linear modeling analysis.
Changes in both starting dosage and dosing requirements
over time were significantly (p<0.001) greater in the
cisatracurium-treated patients.

828
PHARMACOTHERAPY Volume 22, Number 7, 2002
cisatracurium (0.17 µg/kg/min) still increased at
a significantly (p<0.001) greater rate over time
than that for pancuronium (-0.04 µg/kg/min).
Further review of these six patients compared
with the rest of the cisatracurium group failed to
identify factors that could explain the large
increases in dosage required by these patients.
We were surprised to find that pancuronium
dosing decreased over time even though
tachyphylaxis has occurred with pancuronium
administration.8, 9 However, pancuronium
tachyphylaxis has been reported in patients with
either burns or another clearly identifiable cause
for tachyphylaxis (e.g., concomitant phenytoin
therapy). Of the pancuronium-treated patients in
our study, only four were treated for thermal
injuries and three were given phenytoin during
their treatment course. A number of factors may
affect NMBA dosing requirements. To attribute
tachyphylaxis to any observed increase in
cisatracurium or pancuronium infusion rates
over time, all confounders that may affect NMBA
dosing requirements must be considered. We
systematically identified factors that affect NMBA
dosing and recorded each as a possible time-
invariant or time-variant confounder.
We can hypothesize that more severe illness
(calculated using APACHE II) may necessitate
higher NMBA infusion rates.26 Administration of
pressure control ventilation to prevent baro-
trauma in critically ill patients is increasing.29
Many patients undergoing this mode of
ventilation require NMBA therapy. Patients with
hepatic insufficiency have altered cisatracurium
pharmacokinetics, such as lower clearance,
volume of distribution, and time to maximal
effect.5 Renal insufficiency may decrease
pancuronium clearance and lead to drug
accumulation over time and thus lower
pancuronium dosing requirements.2 Tachyphylaxis
related to neuromuscular blockers may occur
more often in patients with burns because of
increased protein binding and upregulation of
acetylcholine receptors.23, 30
neuron lesions (e.g., Guillain-Barré syndrome,
muscular dystrophy), myasthenia gravis, Eaton-
Lambert syndrome, peripheral neuropathy, spinal
cord injury, sepsis, and cancer, may affect
neuromuscular transmission.20, 31, 32, 39, 40 Any of
these disorders might affect NMBA dosing
requirements by either enhancing or diminishing
nerve impulse transmission by way of
upregulation of acetylcholine receptors and
alterations in protein binding.31, 32
Serum potassium and magnesium concentration
and arterial pH affect response to NMBA
therapy.36 Temporal changes in any of these
cation concentrations, therefore, could affect
cisatracurium or pancuronium infusion rates.
Peak airway pressure exceeding 45 mm Hg is a
common reason for beginning NMBA therapy.
Severe hypoxemia, identified by a low PaO2:FiO2
ratio, may warrant administration of NMBA
therapy. Sedation and narcotic analgesia therapy
was evaluated over time because an inverse
dosing relationship with NMBA may exist.
Inadequate sedation or narcotic analgesia therapy
could increase NMBA dosing requirements and
thus mimic tachyphylaxis. However, this factor
probably did not significantly affect our analysis,
because the number of patients who did not
receive concomitant sedation or narcotic
analgesia therapy was equal between groups (two
patients/group), and the infusion rate increase
did not exceed the average daily increase (0.39
µg/kg/min) in either of the two cisatracurium-
treated patients.
Train-of-four monitoring with peripheral nerve
stimulation is routine for patients receiving
continuous NMBA therapy, and infusions usually
are suspended if no twitch is detected. The
number of times an NMBA infusion was
suspended for this reason differed significantly
(p<0.001) between patients treated with
pancuronium (15 episodes) and cisatracurium
(10 episodes). A theoretical concern is that
administering pancuronium as a continuous
infusion could lead to accumulation and thereby
obscure pancuronium-related tachyphylaxis.
However, the likelihood of this sequelae is small,
because the nurse would suspend the
pancuronium infusion if no twitch was detected
with train-of-four monitoring. Institutional
guidelines recommend that NMBA infusions be
suspended until the desired number of twitches
are observed and then reinstituted at a lower
infusion rate.
Several drugs, such as halogenated anesthetics,
local anesthetics, aminoglycosides, clindamycin,
vancomycin, antidysrhythmics, steroids, lithium,
xanthine derivatives, dantrolene, cyclosporine,
anticonvulsants, and anticholinesterase agents,
may affect neuromuscular transmission and
therefore have the potential to affect NMBA
infusion requirements.31–38 Several neurologic
and systemic diseases, such as upper motor
neuron lesions (e.g., multiple sclerosis, cerebral
palsy, amyotrophic lateral sclerosis), lower motor
We observed that cisatracurium was prescribed
more frequently by the medical ICU service and

TACHYPHYLAXIS ASSOCIATED WITH CONTINUOUS CISATRACURIUM THERAPY Kanji et al 829
years.
pancuronium more frequently by the surgical
ICU service. This is probably because patients
with renal insufficiency and those receiving
concomitant corticosteroid therapy (two of the
institutional criteria for cisatracurium therapy)
are more likely to be admitted to the medical
ICU. Five patients in the cisatracurium group
had renal insufficiency (four of these were
admitted to the medical ICU), and two patients
received concomitant corticosteroid therapy
(both were admitted to the medical ICU). We
were surprised that two of the patients with renal
insufficiency received pancuronium therapy
despite institutional guidelines stating that
cisatracurium is the agent of choice for these
patients. Medical records of these two patients
revealed no evidence that prolonged paralysis or
muscle weakness had occurred after
pancuronium was discontinued. Although
patients in the medical ICU are more likely to
have many medical problems, a higher
probability of dying, and higher NMBA
requirements during the terminal stages of
critical illness, no differences were observed in
APACHE II scores, length of stay, or survival
between groups. It is also unlikely that
cisatracurium or pancuronium would be
continued beyond the point at which palliative
treatment would be considered. Last, we
suspected that patients in the surgical ICU were
more likely than patients in the medical ICU to
undergo surgical procedures involving general
anesthesia and NMBA therapy. However, we
determined that none of the patients in the
surgical ICU underwent a surgical procedure
while on continuous NMBA therapy.
The economic impact of these results is
substantial and directly related to the cost of
cisatracurium, which is more than 10 times that
of pancuronium on a milligram-potency basis.
Our study results further justify the
administration of pancuronium as first-line
NMBA therapy at our institution, unless safety
concerns exist (e.g., renal insufficiency or
concomitant steroid administration). Our cost
comparison for patients treated with
cisatracurium versus pancuronium considered
only the NMBA acquisition cost. Although no
patients in our study had documented evidence
of prolonged muscle weakness, this phenomenon
occurs in 5–10% of patients receiving continuous
NMBA therapy and costs approximately
$60,000/episode.41
Conclusion
Tachyphylaxis is common among patients
treated with neuromuscular blockers. The
frequency of tachyphylaxis may be more
common with the benzylisoquinolinium class of
NMBA agents than with the aminosteroid agents.
Our study supports the notion that tachyphylaxis
can occur with continuous cisatracurium therapy.
Further research should be directed toward
identifying the mechanism behind this resistance
and identifying patients who may be at risk for
its development. Clinicians who encounter
tachyphylaxis in patients receiving continuous
cisatracurium therapy should consider
administering a different NMBA agent.
References
A premise of our study was that NMBA therapy
was administered effectively in all instances as
rapidly as possible to meet the clinical goals for
its administration. Slower than desirable
titration of NMBA therapy to clinical effect may
have mimicked tachyphylaxis in some instances.
However, the acute nature of the ICU setting
coupled with the ease of titration of these agents
suggests that this scenario was unlikely. Our
study may not have had the power to identify
risk factors for tachyphylaxis with cisatracurium
therapy because tachyphylaxis with NMBA
administration occurs relatively infrequently.
Safety concerns regarding NMBA therapy have
led to a decrease in its administration over the
past decade.41, 42 Despite the large critical care
population at our institution, we identified only
45 patients who received continuous NMBA
therapy for more than 48 hours over the last 2
1. Coursin DB, Prielipp RC. Use of neuromuscular blocking
drugs in the critically ill patient. Crit Care Clin
1995;11:957–81.
2. Murray MJ, Cowen J, DeBlock H, et al. Clinical practice
guidelines for sustained neuromuscular blockade in the adult
critically ill patient. Crit Care Med 2002;30:142–56.
3. Glaxo Wellcome. Nimbex (cisatracurium besylate) product
information. Research Triangle Park, NC; 2000.
4. Bryson HM, Faulds D. Cisatracurium besylate: a review of its
pharmacology and clinical potential in anesthetic practice.
Drugs 1997;53:848–66.
5. De Wolf AM, Freeman JA, Scott VL, et al. Pharmacokinetics
and pharmacodynamics of cisatracurium in patients with end-
stage liver disease undergoing liver transplantation. Br J
Anaesth 1996;76:624–8.
6. Cardinale V, ed. Drug topics red book. Montvale, NJ: Medical
Economics Company, 2000.
7. Lefkowitz RJ, Hoffman BB, Taylor P. Neurotransmission. In:
Hardman JG, Limbird LE, Molinoff PB, et al, eds. Goodman and
Gilman’s the pharmacological basis of therapeutics, 9th ed.
New York: McGraw-Hill, 1996;105–39.
8. Yamashita M, Shiga T, Matsuki, Oyama T. Unusual resistance
to pancuronium in severely burned patients: case reports. Can
Anaesth Soc J 1982;29:630–1.

830
PHARMACOTHERAPY Volume 22, Number 7, 2002
9. Hickey DR, Sangwan S, Bevan JC. Phenytoin-induced
resistance to pancuronium. Use of atracurium infusion in
management of a neurosurgical patient. Anaesthesia
1988;43:757–9.
II: a severity of disease classification system for acutely ill
patients. Crit Care Med 1985;13:818–29.
27. Cockcroft DW, Gault MH. Prediction of creatinine clearance
from serum creatinine. Nephron 1976;16:31–41.
10. Uyar M, Hepaguslar H, Ugur G, Balcioglu T. Resistance to vecuro-
nium in burned children. Paediatr Anaesth 1999;9:115–18.
11. Buggy D, Murphy K, Van Haaster J, Feely J, Pasureau F.
Resistance followed by prolonged neuromuscular block with
vecuronium. Acta Anaesthesiol Scand 1996;40:862–3.
12. Norman J. Resistance to vecuronium. Anaesthesia 1993;
48:1068–9.
13. Fish DN, Singletary TJ. Cross-resistance to both atracurium-
and vecuronium-induced neuromuscular blockade in a
critically ill patient. Pharmacotherapy 1997;17:1322–7.
14. Kumar CM, Lawler PG. Phenytoin-induced resistance to
vecuronium. Anaesthesia 1989;44:263–4.
15. Hepaguslar H, Uyar M, Ugur G, Balcioglu T. Resistance to
vecuronium after immobilization. Paediatr Anaesth
1999;9:94–5.
16. Moorthy SS, Krishna G, Dierdorf SF. Resistance to
vecuronium in patients with cerebral palsy. Anesth Analg
1991;73:275–7.
28. Bryk A, Raudenbush S. Hierarchical linear models. Newbury
Park, CA: Sage, 1992.
29. Gillette MA, Hess DR. Ventilator-induced lung injury and the
evolution of lung-protective strategies in acute respiratory
distress syndrome. Respir Care 2001;46:130–48.
30. Marathe PH, Dwersteg JF, Pavlin EG, Haschke RH, Heimbach
DM, Slattery JT. Effect of thermal injury on the pharmaco-
kinetics and pharmacodynamics of atracurium in humans.
Anesthesiology 1989;70:752–5.
31. Martyn JA, White DA, Gronert GA, Jaffe RS, Ward JM. Up-
and-down regulation of skeletal muscle acetylcholine receptors:
effects on neuromuscular blockers. Anesthesiology
1992;76:822–43.
32. Tschida SJ, Graupe KJ, Hoey LL, Vance-Bryan K. Resistance to
nondepolarizing neuromuscular blocking agents.
Pharmacotherapy 1996;16:409–18.
33. Kobayashi O, Ohta Y, Kosaka F. Interaction of sevoflurane,
isoflurane, enflurane, and halothane with non-depolarizing
muscle relaxants and their prejunctional effects at the
neuromuscular junction. Acta Med Okayama 1990;44:209–15.
34. Dreissen JJ, Wuis EW, Gielen MJ. Prolonged vecuronium
neuromuscular blockade in a patient receiving orally
administered dantrolene. Anesthesiology 1985;62:523–4.
35. Dehpour AR, Samadian T, Roushanzamir F. Interactions of
aminoglycoside antibiotics and lithium at the neuromuscular
junctions. Drugs Exp Clin Res 1992;8:383–7.
17. Gronert GA. Resistance to atracurium [letter]. Acta
Anaesthesiol Scand 1997;41:327.
18. Brett RS, Schmidt JH, Gage JS, Schartel SA, Poppers PJ.
Measurement of acetylcholine receptor concentration in
skeletal muscle from a patient with multiple sclerosis and
resistance to atracurium. Anesthesiology 1987;66:837–9.
19. Tempelhoff R, Modica PA, Jellish WS, Spitznagel EL.
Resistance to atracurium-induced neuromuscular blockade in
patients with intractable seizure disorders treated with
anticonvulsants. Anesth Analg 1990;71:665–9.
36. al Ahdal O, Bevan DR. Clindamycin-induced neuromuscular
blockade. Can J Anaesth 1995;42:614–7.
20. Tatman AJ, Wrigley SR, Jones RM. Resistance to atracurium in
a patient with an increase in plasma alpha 1 globulins. Br J
Anaesth 1991;67:623–5.
37. Huang KC, Heise A, Shrader AK, Tsueda K. Vancomycin
enhances the neuromuscular blockade of vecuronium. Anesth
Analg 1990;71:194–6.
21. Kelly KJ, Clarens DM, Kohls PR, Nahum A, Vance-Bryan K.
Response to pancuronium after loss of atracurium-induced
neuromuscular blockade. Ann Pharmacother 1993;27:862–5.
22. Luleci N, Akturk G, Kalac N. Resistance to atracurium:
Wegener’s granulomatosis: a case report. Middle East J
Anesthesiol 1994;12:511–15.
23. Dwersteg JF, Pavlin EG, Heimbach DM. Patients with burns
are resistant to atracurium. Anesthesiology 1986;65:517–20.
24. Tschida SJ, Hoey LL, Nahum A, Vance-Bryan K. Atracurium
resistance in a critically ill patient. Pharmacotherapy
1995;15:533–9.
38. Ganjoo P, Tewari P. Oral cyclosporine-vecuronium interaction
[letter]. Can J Anaesth 1994;41:1017.
39. Nilsson E, Meretoja OA. Vecuronium dose-response and
maintenance requirements in patients with myasthenia gravis.
Anesthesiology 1990;73:28–32.
40. Knuttgen D, Jahn M, Zeidler D, Doehn M. Atracurium during
thoracic surgery: impaired efficiency in septic processes. J
Cardiothorac Vasc Anesth 1999;13:26–9.
41. Rudis MI, Guslits BJ, Peterson EL, Hathaway SJ, Angus E,
Beis S. Economic impact of prolonged motor weakness
complicating neuromuscular blockade in the intensive care
unit. Crit Care Med 1996;24:1749–56.
25. Baraka A, Idriss N. Resistance to rocuronium in an epileptic
patient on long-term carbamazepine therapy: a case report.
Middle East J Anesthesiol 1996;13:561–4.
42. Lumb PD. Economic impact of prolonged neuromuscular
weakness complicating neuromuscular blockade in the
intensive care unit. Crit Care Med 1996;24:1615–16.
26. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE